California Public Sector Retirement Programs and Compensation

California Public Sector Retirement
Programs and Compensation

A financial analysis prepared for
The California Foundation for Fiscal Responsibility

Capitol Matrix Consulting
July, 2011

This page is intentionally blank for printing purposes
Table of Contents
Executive Summary ....................................................................................................... 1
Chapter 1: Retirement Programs ................................................................................. 7
Appendices: ......................................................................................................................... 45
Chapter 2: Compensation .......................................................................................... 73
Appendices ........................................................................................................................... 87
Chapter 3: Government Cost of Retirement Programs ........................................... 107
Appendices ......................................................................................................................... 129



Executive Summary 1
Executive Summary

California state and local pension systems face major financial shortfalls, which have resulted
from retroactive benefit enhancements, shortfalls in investment returns, greater-than-expected
wage increases, and other experience less favorable than had been assumed. The State and many
local governments also face major looming increases in retiree health care costs as baby-boomers
retire, reflecting the fact that almost no funds for this benefit are being set aside in advance.
This report is prepared in response to a request by The California Foundation for Fiscal
Responsibility (CFFR) for an analysis of retiree benefits and taxpayer costs for California state
and local government retirement programs. We also look at the impact of two alternative CFFR
reform proposals.
CFFR Alternatives
This report estimates the impact that two CFFR reform proposals would have on benefits
received and on taxpayer costs.
The first reform, Alternative A, provides that current employee pension contributions must
immediately increase as necessary to cover one-half of the expected cost of additional accruals. It
also provides that future hires are to be covered by a retirement income program that is no more
generous than a modified version of the pension and thrift savings plans that currently apply to
federal employees; the modifications are a wage cap under the pension component (with an
additional employer contribution for pay in excess of the cap under the thrift plan), and a
requirement that employees pay at least one-half of the expected cost of future pension accruals.
Retiree healthcare benefits for future hires are to be no more generous than per the program
currently applicable to state employees, modified to include significantly reduced employer
subsidies. Finally, Alternative A would prospectively apply the full restrictions applicable to future
hires to current employees upon declaration of a Fiscal Emergency by the relevant governmental
entity. We modeled Alternative A by assuming that the most generous permitted provisions would
apply. For a scenario where the reform would apply to current employees, we assumed that the
Fiscal Emergency would be effective in early 2013.
The second proposal, Alternative B, limits employer contributions to cover the cost of pension or
defined contribution benefits earned by current or future employees for service after June 2012 to
6 percent of payroll (9 percent for safety classifications). We modeled implementation of this in
terms of a dollar-for-dollar matching arrangement within a defined contribution plan. This
reform does not address retiree healthcare benefits.
Both Alternatives provide for additional Social Security replacement benefits on a pension basis
for employees not covered by that federal program.
!

!

2 Executive Summary
Chapter 1
The first chapter compares the value of employer-provided retirement benefits among major
California public sector groups with the value of the benefits provided by the federal government,
and by a sample of large private sector employers in the state. It compares lifetime employer-
provided benefits in present value terms, which converts future streams of payments into todays
(2011) dollars. Such a methodology enables us compare different types of retirement programs
(such as pensions and defined contribution programs) on a comparable basis.
The chapter also looks at the impact that the two CFFR reform alternatives would have on these
comparisons for employees terminating service at different ages with varying levels of service and
salary. The California public sector groups we cover are: state miscellaneous (non-safety)
employees; (2) CHP employees; (3) public school teachers; (4) non-safety employees of a
representative local system.
Key findings for Chapter 1 are:
Non-safety state employees. For most full-career employees, public sector systems in
California provide retirement income benefits that are moderately larger than the federal
government provides, and substantially larger than what is available within the private
sector. For state members covered by the CalPERS system, new employees hired after
early 2011 receive benefits that are less than those hired before that date, but the benefits
levels remain significantly above that of the private sector. Finally, employees who do not
remain within state and local retirement systems for an entire career would sometimes do
better in private sector programs.
Teachers. Retirement benefits received by this group are significantly less generous
than most other public sector employees. This is partly because teachers covered by
CalSTRS are not in social security, their retiree health care is provided by school districts
and tends to be less generous than the State of California, and their pension formulas for
those terminating before full retirement age is less generous than other public funds.
Safety employees. Safety employees receive much higher benefits than other employee
groups. Compared to their federal counterparts, the comparisons are mixed. Our
modeling of the system covering CHP members indicates that those hired before mid-
2010 receive pension and retiree health benefits that are higher than their federal law
enforcement counterparts. Those hired after that date, however, have benefits that are
similar to federal law enforcement employees. We did not compare these benefits to the
private sector, since there are no direct counterparts.
Local governments. Though there is considerable variation, a number of local
governments offer pension benefits that are significantly higher than the state. This
reflects more generous benefit formulas for early retirement and employer pick-up of
employee contributions.
Retiree healthcare. This is a major benefit found in the public sector particularly at
the state level -- that has become increasingly rare in the private sector. Given the rise in
healthcare costs, this benefit is often more valuable than the pension earned by low- and
moderate-income employees.


Executive Summary 3
CFFR reforms. In general, the two CFFR reforms would significantly reduce benefits
earned by full-career California public sector employees (a key exception being teachers,
who would, in some cases, receive modestly higher benefits), but would leave benefits
significantly above private sector levels. Also, both reform Alternatives would increase
benefits for those who work in the California public sector only at the earlier stages of
their career. Finally, both alternatives shift risks associated with investment shortfalls and
other factors from the employer to the employee.
Chapter 2

The second chapter broadens the discussion by comparing total compensation (wages and
benefits) earned by employees in state and local governments with compensation earned for
comparable jobs in the private sector. We develop information from wage surveys by the Bureau
of Labor Statistics and the State Department of Personnel Administration, statistical studies using
Current Population Survey data developed by the U.S. Census Bureau, and other sources.
Among our key findings:
Wages non-safety employees. Average wages in the state and local government
sector combined are roughly similar to average private sector pay levels for comparable
jobs a little above the average for all private sector workers, and a little below private
sector workers employed by large firms. Relative to the private sector, public sector pay
tends to be higher in less-skilled occupations, and sometimes lower in more specialized
and high skilled occupations. Within occupations, there is much greater pay variability in
the private sector. We also discuss the significant limitations that these occupational
surveys and statistical studies have.
Wages safety employees. Public-private sector comparisons for safety employees is
complicated by the fact that peace officers, firefighters, and correctional officers do not
have direct private sector counterparts. Compared to safety employees employed by
public agencies in other states, we find that pay rates in California are above average.
Non-wage benefits. The incidence of non-wage benefits is higher in the state and local
government sector than in the private sector. Moreover, when compared to the subset of
private employers that offer a full range of benefits, public sector benefits are more
generous. The main difference is in retirement benefits, where long-term employees of
state and local governments enjoy pension and retiree healthcare benefits that are far
richer than what is generally available within the private sector.
Total compensation. Given that wages for comparable occupations are similar, the
higher non-wage benefits can raise total compensation for a typical full-career public
sector employee significantly above his or her private sector counterpart, particularly for
public sector employees who receive both pension and retiree healthcare benefits. Again,
these findings reflect averages. Given the greater variation in private sector pay levels, the
findings for typical employees may or may not hold for individuals in specific
occupations and job levels.
!

!

4 Executive Summary
Chapter 3
The third chapter compares the costs that California government agencies incur to provide
retirement benefits under current law with costs under the CFFR Alternatives. We develop
detailed estimates for the groups covered in Chapter 1 (state non-safety and CHP employees,
public school teachers, and several hypothetical local government groups), using the actuarial
assumptions adopted by the governing bodies for each system. From these detailed calculations,
we also draw more general inferences about the potential cost impact of implementing the
alternatives across all state and local retirement funds. We also look at the impact on relative costs
of alternative rates of pension investment return.
Our analysis focuses on the actuarially determined costs under current law and the Alternatives
the amounts that employers should be setting aside to fund accruals of retirement benefits. These
costs provide a good indication of the ultimate impacts of the reforms that we model. However, as
discussed more fully in the chapter, their impact on near-term budget outlays depends on the
funding policies of the programs involved. In the majority of cases, changes in actuarially
determined retirement program costs have immediate effects on budget outlays, but there are
exceptions. In particular, since governments generally do not prefund retiree health benefits,
reforms reducing costs for these programs would not impact budget outlays until the affected
employees retire and start drawing benefits sometimes decades in the future
Current Costs
Based on projections of historical data reported to the State Controllers Office, we estimate that
State and local governments currently contribute about $16 billion per year to cover the cost of
public pensions in California. Employee pension contributions amount to an additional $11
billion, and other costs are a bit over $1 billion. (Employer costs may be understated, and
employee costs similarly overstated, to the extent that employer pickup of employee contributions
is not fully reflected in the data being reported to the Controller.) In addition, state and local
governments are accruing obligations for retiree health benefits at the rate about $5 billion per
year although most of the budget impact for these costs is delayed, since they are not being
prefunded.
Impact of Reforms
When fully implemented, the CFFR alternatives would reduce employer costs at the state and
local level. Our key findings are:
State-level savings. Using existing actuarial assumptions, Alternative A would result in
a net reduction in employer costs for retirement income benefits in the low hundreds of
millions of dollars per year. The reduction in annual costs for retiree healthcare would be
about $2 billion per year. Under Alternative B, costs for retirement income programs
would be about 4 percent of payroll reduction in annual pension cost ($2.7 billion in
todays dollars). This alternative does not affect the retiree health benefit.


Executive Summary 5
Factors behind results. The relatively modest change in state costs for retirement
income programs, particularly with respect to Alternative A, is partly the result of recent
actions taken by the state that is resulting in lower CalPERS costs under existing law.
These include increased employee contributions, and lower benefits for new employees.
For the large non-safety group and teachers, the cost of the combined pension and
defined contribution program that we modeled under Alternative A is slightly higher than
the cost for current and future hires under existing law. Also, Alternative A was
modeled assuming implementation of a modified version of the federal employee thrift
plan provisions, with an expected employer cost of about 5 percent of payroll. However,
this alternative would also accommodate a less-generous defined contribution design for
example, where the employer contribution was limited to 3 percent of pay. Each one
percent reduction translates into about $700 million in savings at the state level.
Impact of lower investment returns. In addition to their impact on expected costs,
the reforms would substantially reduce the risk of state and local governments facing
additional strains on their budgets from less-favorable-than-expected experience related
to future investment returns and other factors. Savings under the alternatives expand
markedly when the assumed rate of future pension investment return is lowered from the
current 7.75 percent per year as is urged by some critics. For example, if we assume
only a 5.75 percent rate (generally consistent with federal law governing private sector
pension sponsors), the annual savings from the pension reforms expand to over $3 billion
under Alternative A and over $7.5 billion under Alternative B. The savings occur because
both alternatives rely on defined contribution programs for part or all of the retirement
income benefit, and under these plans employer costs are unaffected by future investment
return.
Local government savings. Under Alternative A, average local government
retirement costs would fall by 5.5 percent to 7.5 percent of payroll when fully phased in
or about $3 billion to $4 billion annually in todays dollars. Adoption of Alternative A
health reforms would produce additional savings (potentially in the high hundreds of
millions annually). Under Alternative B, the savings would be would be 7.5 percent to 9
percent of payroll, or about $4 billion to $5 billion in todays dollars.
Factors behind results. The relatively larger savings compared to the state is related
to the richer plan designs and lower employee contributions at the local level. As with the
state-level comparisons, the anticipated savings under the alternatives increase markedly
as we lower the expected rate of pension investment return from the currently assumed
7.75% per year.
Timing of savings. If the programmatic reforms applied to future accruals of existing
employees, the savings for the retirement income programs identified above would occur
immediately. Given the lack of prefunding, retiree health care savings would occur only
after the existing employees retired. If the reforms applied only to future employees, the
savings from the programmatic changes would emerge gradually over time as the existing
workforce turns over. (In the case of retiree health, the savings would occur when the
newly hired employees themselves retire.) However, under Alternative A, the
requirement that all employees pay one-half of the normal costs of their pensions would
result in immediate local savings of over $2 billion at the local level. (State cost reductions
would be minor since employees are now paying nearly half of normal costs in many
cases). And, if the employer contribution caps under Alternative B (6 percent of pay for
non-safety employees and 9 percent for safety employees) were applied to existing pension
systems (as opposed to just those covered by new defined contribution systems), most of
the above-identified savings for this alternative would occur immediately.
!

!

6 Executive Summary
Conclusion
Implementation of both CFFR reforms would result in significant savings at the state level, and
relatively large savings at the local level. The timing and magnitude of the savings would
depend on how the reforms were implemented. Just as important as impact on expected costs, the
reforms reduce governmental (and taxpayer) exposure to major future cost increases due to less-
favorable-than-expected outcomes relating to investment returns and other factors.

Chapter 1: Retirement Programs 7

!
Chapter 1: Retirement Programs
Introduction
Public pension funds in California face massive shortfalls, which have resulted from retroactive
benefit enhancements, as well as recent investment returns and other experience less favorable
than assumed. The combined actuarial shortfalls for the states ten largest public funds, which
account for 90 percent of public pension fund assets in the state, is $240 billion, or nearly one-
third the combined liabilities of the funds
1
. Using discount rate assumptions consistent with
federal mandates for private sector funds, the shortfalls are much larger.
The state and many local governments will also face major increases in retiree health care
demands as baby-boomers retire. While the problem with respect to pensions is that there are
inadequate funds relative to whats been promised, the problem with retiree health care is starker:
little or no money has been set aside for benefits that, for some members, are almost as significant
as their pension. These rising costs and consequent fiscal pressures, coupled with a growing
disparity between public and private sector retirement benefits, have generated considerable
debate about reform.
Purpose of This Report
This report is in response to a request by The California Foundation for Fiscal Responsibility
(CFFR) for a financial analysis of retiree benefits and taxpayer costs for California state and local
government retirement programs both under existing law and under two of its proposed
alternatives.
The first chapter of this report discusses recent developments relating to the provision of
retirement benefits in the public and private sector, and then turns to an analysis of these benefits.
Its purpose is twofold:
To show how retirement benefits in Californias state and local government sector compare to
those in the private sector and under the federal system, using sample employees in a variety of
circumstances.
To show how the retirement benefits of these employees would be affected by the CFFR
proposed pension reforms.
Our comparisons are based on detailed modeling of benefits received by four California public
sector employee groups
2
, employees of the federal government, and California private sector
employees. We also model benefits received under the CFFR proposals, described as:
Alternative A, which includes a modified version of the pension and thrift savings plans that
apply to federal employees, and separate reforms to the current state retiree health benefit
program.

1
Source: Little Hoover Commission. Public Pensions For Retirement Security. February 2011.
http://www.lhc.ca.gov/studies/204/Report204.pdf
2
Specifically, non-safety (miscellaneous) state employees, CHP employees, public school teachers, and non-
safety employees of a sample local system. CalPERS administers pension and retiree health benefits for
state non-safety and CHP employees, and pension benefits for our sample local entity. CalSTRS provides
pension benefits for teachers. Retiree health benefits are separately provided by the local entity for non-
safety local employees, and by the employing school district for teachers.
!
8 Chapter 1: Retirement Programs

Capitol Matrix Consulting !
Alternative B, which limits employer contributions to cover the normal cost for future
retirement income benefits (excluding retiree health care) to 6 percent of payroll (9 percent for
safety classifications). It provides an additional benefit for those not covered by Social Security,
but does not otherwise mandate a specific benefit design, and does not address retiree health
care benefits. For purposes of this chapter, we model future benefits (other than the Social
Security replacement benefit, if any) via a defined contribution plan where employers match
employee contributions on a dollar-for-dollar basis, up to 6 percent of pay (9 percent for safety
classifications).
The current programs and alternative reforms are discussed in more detail below.
The subsequent chapters of this report address the broader question of how total compensation
(wages and benefits) in the state and local sector compares to the private sector, and provide
estimates of the impacts of CFFR alternatives on contributions to be made by public employers
and employees in the future.
Report is a Financial Analysis
The CFFR alternatives considered here would impact retirement benefits earned by California
state and local government employees for service after the reform effective date, including current
employees. Applying the reforms to current employees as well as to future hires is necessary in
order to provide governments and taxpayers with meaningful cost relief over the coming decade.
Although not modeled here, it might also be appropriate to apply at least some elements of retiree
health care reform to those already retired. However, it is important to note that this report is a
financial analysis, and leaves to others the critical matter of the extent to which application of
these reforms to current program members would be legally feasible.
Key Findings
Our key findings with regard to existing California public sector programs are as follows:
1) California public sector retirement programs are much richer than those in
the private sector for full-career employees. Members in three of the four
California public employee groups we modeled receive employer-funded benefits that are
considerably larger than those provided through private sector plans offered by our sample
companies often two to three times the private sector benefit level. (This includes CHP
employees, although our results do not explicitly compare their benefits with those
provided to private sector employees.) The exception is teachers, who receive benefits that
are only modestly larger.
2) They can be less generous for short-term employees. In some scenarios involving
members who terminate employment prior to minimum retirement age, the benefits
accrued under the public systems are less than in the private sector. This is a characteristic
of pension plans generally, where the value of benefit accruals accelerates rapidly during
the latter stages of a career; private sector plans rely much more heavily on defined
contribution or hybrid defined benefit plans, where benefits accumulate more
proportionally through a career. In fact, at the youngest ages an employees contribution to
the public sector plans considered here is worth more than the pension being earned. This
result also reflects the fact that no retiree health care benefits are provided for termination
before earliest retirement age.


!
3) The retiree health subsidy is a large benefit. For a full career state employee,
depending on age at retirement and other factors, the value of the retiree health care
benefit can be several hundred thousand dollars. This partly reflects the expectation that
health care cost increases will continue to outpace general inflation. The value of this
benefit alone can exceed the value of the total retirement benefits that would be provided
under private sector plans, which typically do not promise any future retiree health care
benefit to current employees. For some low and moderate wage state employees, the retiree
health benefit is worth more than the pension. This is because, in contrast to the pension
benefit, the retiree health benefit is not related to wages. The value of this benefit is less
dramatic, though still potentially large, for employees of many local governments and
school districts, due to typically less generous post-retirement subsidies.
4) Comparisons to the federal retirement system are mixed. Full career employees
of the state and the representative local system generally receive benefits that are
comparable to or larger than the benefits received by federal employees. (In the case of
state employees, the exact result depends on the time of hire.) In contrast, teachers retiring
early earn smaller benefits under current law than they would under the federal retirement
program.
5) Teachers earn less generous benefits than other California public employees.
Teachers receive smaller benefits than the other California public sector employees
considered here. The main reasons are (1) the pension formulas have relatively steeper
payment reductions for those retiring early, (2) employer-provided health care is
determined by the district, and on average is less generous than the state provides, and
(3) teachers have been required to make larger pension contributions than others relative to
the benefits they receive (this is especially true for past years). In addition, teachers, like
CHP employees but unlike other employees considered in this report, do not participate in
social security.
6) Funding risks are largely borne by employer in California public sector
funds. In each of the public sector examples, all of the risks associated with investment
returns and life span, and much of the risk posed by inflation, are borne by the employer
(and ultimately the taxpayer). This is in contrast to the federal system and, particularly, the
private sector systems, which leave some or most of this risk with the employee.
Our Key Findings with respect to the CFFR alternatives are as follows:
1) CFFR alternatives provide a moderate level of benefits. For full career employees
(for example, entering service at age 27 and retiring at age 57), the benefits are still much
larger than average private sector benefits in several examples, about double. In some
cases, the alternatives would increase benefits for teachers. However, for state employees,
the benefits received under the alternatives are about 25 percent smaller than what is
provided to those hired after early 2011, and about 40 percent less than is provided to
those hired before 2011. For different reasons, each alternative produces moderately
smaller benefits than the federal system.
2) Alternatives offer similar levels of benefits. Even though the two CFFR alternatives
differ a great deal in their details, they produce similar levels of benefits for employees in a
wide variety of circumstances.
!

3) Impacts of alternatives on existing employees vary. As noted above, results for
mid-career employees are based on a blend of pension accruals occurring under existing
law up to the operative date of the alternatives, and generally lower accruals under the
alternative reforms thereafter. Not surprisingly, our comparisons show that the pension
impact of these changes would be minor for late career employees, but more pronounced
for younger employees, who would accrue benefits for more years after the changes are
implemented. The reductions on total retirement benefits would be somewhat more
significant under Alternative A for employees retiring before age 62, because that reform
would scale back early retiree health care subsidies currently provided.
4) Both alternatives shift funding risk to employees. As noted above, under the
existing California public sector systems included in our comparisons, investment, longevity
and a large part of inflation risk are borne by the employer (taxpayers). Over time, both
alternatives share these risks with the employee.
5) Alternative proposals would have small effect on expected benefits for
teachers. Both Alternative A and Alternative B include a provision for social security
replacement for members not currently in the social security system, including teachers and
CHP employees. Net changes in benefits accruing to teachers under the alternatives would
not be large, and in some cases members would receive more total benefits. In any event, at
least some of the investment, longevity and inflation risk would shift to employees under
these alternatives.
6) Alternative A would reduce benefits to younger retirees and those without
long service. This alternative incorporates features of the federal pension system for
future accruals. Unlike the California public sector plans considered here, the federal plan
provides significantly reduced pension benefits to members who leave service before
reaching key eligibility thresholds (for example, age 57 with 30 years of service, or age 60
with 20 years; different thresholds apply for law-enforcement and other safety employees).
Alternative A also reduces employer health subsidies for early retirees. Provisions
discouraging early retirement should lower employer benefit costs, although the impact on
total payroll costs and workforce management should also be considered.
7) Alternative A wage cap has dual effects. A provision in alternative A limits the
amount of annual earnings recognized for future pension accruals to about $80,000
(adjusted for inflation). Income above this cap is recognized for a special employer
contribution under the defined contribution component, but the benefits arising from this
contribution wont fully make up for the caps impact on a retirees pension. Our scenarios
suggest the cap has a moderate effect on overall benefits except for the most highly
compensated employees. Its main effect is that over time, highly compensated employees
would start to bear more investment, longevity and inflation risk than other employees.
In the subsequent sections of this chapter, we provide background on public pensions in
California, discuss recent public sector and private sector developments in the retirement area,
and then turn to our detailed comparisons.
Retirement Systems In California
The California State Controller reports that the state has 131 public retirement systems, including
10 state systems, 20 systems operating under the County Employees Retirement Law of 1937
(1937 Act), 1 independent county system (San Luis Obispo County), 36 city systems, 55 special
district systems, 4 school district systems, and 5 other systems. The remainder of cities, counties,


!
and special districts contract with the California Public Retirement System (CalPERS) to
administer their retirement benefits (some individually and others through pools).
3

Of these 131 systems reported to the State Controller, 85 are defined benefit pension systems and
46 are defined contribution systems. (These figures exclude an unknown but significant number of
supplemental defined contribution plans maintained by local agencies but not reported to the
Controller). Defined benefit plans account for the vast majority of public sector retirement
coverage, whether measured by fund assets or membership. About 81 percent of public
employees in the state are covered through defined benefit plans about 20 percent are covered by
defined contributions. In some cases there is overlap, where the defined contribution plan
supplements the defined benefit program. As shown in Figure 1, members in defined contribution
plans account for marginal or non-existent shares of the total in all categories except special
districts.
Figure 1
Participants in California Public Sector Retirement Programs
(Thousands of Members)

Four-fifths Of California Public Members In 5 Funds
About 80 percent of state and local employees in California are covered through five large
pension funds. As shown in Figure 2, the largest is CalPERS, with 1.6 million members that are
divided roughly equally between State of California members, school district classified members,
and local contracting agency members. The contracting agencies represent over 2,100 cities,
counties, special districts, housing authorities and other governmental entities that contract with
CalPERS to administer their benefits.
The second largest is the California State Teachers Retirement System (CalSTRS), which is a
statewide system that provides benefits to K-12 and community college teachers. The remainder
of the top five consists of the University of California, Los Angeles County, and the City and
County of San Francisco.

3
Source: Public Retirement Systems Annual Report, 2007-08. California State Controllers Office.
http://www.sco.ca.gov/ard_locrep_retirement.html
!

Figure 2
Largest Public Pension Systems In California
(Dollars in billions)
Public Pension System
Recent
Assets
Recent
Funded Status
Public Employees Retirement System $231 61%
State Teachers Retirement System $141 63%
University of California $45 73%
Los Angeles County Employees Retirement Assn. $40 69%
San Francisco City and County Retirement System $16 74%
California Public Pension Funds Face Major Actuarial Shortfalls
The combination of retroactive benefit enhancements, lengthening lifespan, and some weaker
than expected investment returns during recent years has led to deterioration in the condition of
public funds in California. As of 2010, the five largest public pension systems are only between
61 percent and 74 percent funded for benefits earned for past service, based on the actuarial
assumptions used by those systems. These measures would be even lower if they instead reflected
the assumptions mandated for use by private sector plans, or those used by the United States with
respect to federal employee pensions. The funded status amounts also do not reflect liabilities, if
any, for outstanding pension obligation bonds.
Retiree Health Care Costs Also Rising
Retiree health care presents a massive financial challenge. Unlike pensions, which are prefunded
through contributions and investment earnings over the employees working career, retiree health
care has largely been financed on a pay-as-you-go basis. This means that state and local
governments have been making significant benefit commitments without undertaking the current
financial sacrifices needed to meet these commitments.
In 2004, the Governmental Accounting Standards Board issued a directive requiring that state
and local governments report the annual costs of retiree health care on an actuarial basis similar
to that used for pensions. The most recent actuarial valuation for the state of California found
that its retirement health care program has an unfunded liability of $60 billion
4
. The figure
represents the present value of the portion of future retiree health benefits for employees and
retirees that is attributable to their past service, for which no money has been set aside. Earlier
estimates suggest that the total unfunded liability for other public agencies may rival that of the
state of California.
5

4
Source: State of California Retiree Benefits Program, GASB # 43 & 45, Actual Valuation Report as of
June 30, 2010. Prepared by Gabriel Roeder Smith & Co. for the State of California. March 4, 2011. The
valuation also found that the normal cost (that is, the amount it would take to cover just the benefits being
accrued for service during 2011-12) would be $2.2 billion, and the current year cost to amortize the
unfunded liability over 30 years would be an additional $2.5 billion.
5
Local jurisdictions report that they are beginning to prefund systems, but generally at rates that are well
below that needed to make a dent in the large unfunded liabilities. At the state level, the California
Highway Patrol (Bargaining Unit 5) 2008 contract included a small set aside to start prefunding retiree
health benefits. Under the 2010 contract, this set aside is redirected to the CHP pension until 2013, when
contributions will resume at a 4% rate. Even the 4% rate will be only 20% of what would be needed to
meet the actuarially required contribution.


!
Bottom LineMajor Cost Increases Ahead
The pension shortfalls, combined with the impact of rising retiree health care costs, imply sharp
increases in state and local retirement related expenditures over the next decade. Recent estimates
by CalPERS actuaries indicate that state employer contributions will need to increase from
16 percent of payroll in 2009-10 to 25 percent by the middle of the next decade to amortize its
$49 billion state unfunded liability. The increase in future CalSTRS contributions would need to
be even greater to cover its $56 billion funding shortfall. On the health care side, based on
moderate assumptions about retirements and medical price inflation, we estimate that annual
cash flow costs for State of California retirees, which are about $1.4 billion in 2010-11, will
quadruple by the middle of the next decade.
In short, absent significant changes to benefit accruals, public employer obligations for retirement
benefits will rise sharply over the next decade, further squeezing governmental budgets that are
already facing enormous pressures.
Trends In Retirement Coverage
In this section, we review the current status of, and recent trends in, the provision of retirement
benefits in the public and private sector. There are stark differences between the two sectors in
this area:
1) Public sector is more likely to have retirement benefits. According to the BLS
National Compensation Survey, the share of state and local employees with access to
retirement coverage is 92 percent in the Pacific Region of the U.S. (which is dominated by
California) compared to 60 percent for all private sector employees.
6
Retirement coverage
is more prevalent for workers in large firms. The access percentage for firms with more
than 100 employees is about 80 percent.
Figure 3
Percentage of Employees With Access To
Employer-Provided Retirement Coverage
Pacific Region

6
Source: BLS, Employee Benefits Survey, March 2010!

!

2) And it is much more likely to have pensions. About 87 percent of state and local
employees have access to defined benefit plans, versus just 20 percent of private sector
employees in the Pacific Region of the U.S. Furthermore, the private sector trend for this
benefit is downward. Figure 4 shows that participation in defined benefit plans by private
sector employees of medium and large sized companies (those with 100 or more employees)
fell from 37 percent in 1999 to 30 percent in 2010. This trend will continue as new firms
rely on defined contribution plans and existing firms continue to close participation to new
hires or freeze pension accruals for all employees. Nationally, 23 percent of private sector
defined benefit plans in companies with 100 or more employees have been frozen.
7

Figure 4
Recent Trends in Private Sector
Percentage Participation in Defined Benefit Plans

3) Remaining private sector pensions are less generous. Even before the freezes and
benefit reductions, a typical private sector plan was based on less generous formulas, had
steeper earlier retirement reductions, and less than 4 percent promised inflationary
adjustments to pension payments. A BLS survey of private sector defined benefit plans in
the 1990s found that an average plan replaced less than one-third of the salary of a worker
retiring at age 65 with 30 years of service.
8
Under the CalPERS 2 percent at 55 formula,
the replacement rate would be more than twice that level. The nearby box shows the key
elements accounting for the added value of a CalPERS pension, relative to a private sector
design that was typical when private sector pension plans were still common.

7
Source: BLS, Employee Benefits Survey, March 2010.
http://www.bls.gov/ncs/ebs/benefits/2010/benefits_retirement.htm
8
Source: Public and Private Sector Defined Benefit Pensions: In Compensation and Working Conditions,
Summer, 1997. Bureau of Labor Statistics.


!
4) Retiree health benefits are offered to majority of public sector employees.
Although precise figures are not available, it appears that about three-quarters of state and
local government employees in California receive some type of post-retirement health care
benefit. The state and UC offer retiree health coverage comparable to what is offered to
active employees through age 64, and medical supplement plans for retirees age 65
and over.
For teachers, CalSTRS does not provide statewide retiree health care coverage. Instead
coverage for teachers is on a district-by-district basis. Precise data is not available for school
districts but recent surveys found that districts covering well over one-half of the members
provided some type of post retirement health care, though coverage after age 65 is
uncommon. Most local governments offer retiree health care, though the subsidies vary a
great deal.
5) Retiree health benefits are disappearing in the private sector. A key factor
affecting the private sector coverage is accounting rules that began to apply to private
sector employers in the early 1990s. They require companies to charge against earnings the
full cost of an employees expected post-retirement health benefits over his working career,
and to make detailed disclosures about liabilities.
9
According to the Employee Benefit
Research Institute, 22 percent of workers were employed at a private establishment that
offered health benefits to early retirees in 2008, down from 31 percent in 1997, while
17 percent of workers were employed at a private establishment that offered health benefits
to Medicare-eligible retirees, down from 28 percent in 1997. Where offered, the subsidies
provided by private sector retirement plans are much more limited.
Recent Developments
The large pension shortfalls and the threats they pose to public services in California have
generated considerable interest in pension reform. Interest has also been sparked by outrage over
pension abuses, such as extreme pension spiking.
Some actions have been taken. At the state level, bargaining agreements and legislation passed in
2010 created less generous benefit formulas for new employees, and higher contributions for
members covered by CalPERS. These changes are recognized in our comparisons below.
CalPERS reports that after years of benefit enhancements that applied to current workers, local
agencies are starting to adopt changes that reduce benefits for future hires. Local voters also
passed nine local pension reform initiatives in November 2010, and many more are being
considered.
For example, the San Diego mayor and a city councilmember announced a pension reform
measure in April 2011, which would be put on the June 2012 ballot. The measure would require
all new city employees except police officers be provided with a defined contribution plan, and
that existing employees pensionable pay be capped for 5 years. It would also remove special pay
items from eligibility for pension calculations. Reform measures are being developed in other
local communities, including San Jose, Los Angeles, and San Francisco.
One of the questions raised by the local efforts is whether such a piecemeal approach will
adequately address broader Californias challenges. The rationale behind the CFFR proposals is
that a statewide solution is needed for this purpose.

9
Source: Paul Fronstin, Implications of Health Reform for Retiree Health Benefits, EBRI Issue Brief, no.
338 (January 2010). www.ebri.org.
!

Comparing CalPERS to Private Sector Pensions
As the results in this chapter show, California public sector retirement benefits are larger than
those provided by private sector employers. One reason is that pensions are disappearing from
the private sector. Equally important, however, public sector plans are more generous even when
compared with traditional private sector pensions. Consider a state employee retiring at age 55
with 25 years of service, and final years pay of $75,000. The employer-provided portion of his
CalPERS pension has a present value of $527,000.
Under a common private sector design used by employers 20 years ago, this employee would
retire today with a pension worth $163,000, plus accumulated 401(k) matching contributions of
$58,000 a total employer-provided value of $221,000.
What makes the CalPERS pension so much more valuable? The following chart starts with the
private sector pension on the left and builds to the CalPERS value in steps.

Pay averaging. CalPERS uses one-year averaging versus five years under most private sector
plans. This boosts the CalPERS benefit.
Benefit factor. At full retirement age, the factor is 2.5 percent under CalPERS, compared to
1.4 percent under our sample private sector plan.
Early payment. A smaller payment reduction for retiring early in CalPERS 20 percent
reduction at age 55 under CalPERS versus a 50 percent reduction in this private sector plan.
Survivor benefit. A free 25-percent benefit continuation (plus $2,000 lump sum) for survivor
under CalPERS. Typically no comparable benefit in private sector plan.
COLAs. Guaranteed cost of living adjustments are provided by CalPERS but not by private
sector plans.
Employee funding. Employee contributions reduce the net employer-provided value under
CalPERS, but private sector pensions are almost always 100-percent employer funded.
Note that order matters: for example, the pay averaging step would be larger (and the other steps
smaller) if it were considered later rather than first.


!
Little Hoover Commission Report
In February 2011, the Little Hoover Commission released a report that documented these
challenges, provided history on how they developed, and set forth a general approach for dealing
with them. The report asserted that public pension funds are so dangerously underfunded that
aggressive statewide reforms are needed.
To address these problems, the report recommended that the Legislature pursue structural
changes that include, for both state and local governments, (1) a reduction to future pension
accruals of current employees and (2) the development of a hybrid model similar to the Federal
Employees Retirement System. Such a model would contain a reduced defined benefit program
along with defined contribution program. The purpose would be to both lower costs and shift
some of the funding risks to employees. The report also recommends that wages subject to the
defined benefit program be capped, in order to further reduce employer risk. Their report does
not address retiree health care benefits.
The CFFR has put forth two proposals that would implement statewide reform. The CFFR
Alternative A proposal embodies many of the Little Hoover Commissions recommendations.
Alternative B takes a more permissive approach, merely setting caps on the normal costs that
public employers can incur for retirement income benefits.
Pension Systems Modeled In Our Comparisons
Our comparisons are based on the application of the specific provisions of each of the retirement
systems to sample employees. The features of each system and of the alternatives modeled in this
chapter are briefly summarized in the following sections. Additional detail is contained in the
appendix.
State Miscellaneous (Non-Safety) Employees
Pension. A member with five years of service is eligible to draw a pension as early as age 50.
The amount equals the product of years of service, highest average monthly pay rate and a
benefit factor based on age when payments begin. A member in a 2 percent at age 55 formula
retiring at age 55 with 30 years of service and $6,000 final monthly average pay would receive an
initial unmodified allowance of 30 X 2% X $6,000, or $3,600 per month.
Pay is averaged over 12 months if the member was first hired before 2007, and 36 months for
those hired subsequently. For those hired before 2011, the benefit factor ranges from 1.1 percent
at age 50 to 2.5 percent at age 63 or older. Less generous factors apply to those hired thereafter.
A lump sum of $2,000 is paid upon the members death, and 25 percent of the members pension
continues for the beneficiarys remaining lifetime; the members pension is not reduced unless he
or she elects additional survivor protection. There is an annual cost of living increase of up to
2 percent (compounded), and further increases as necessary to maintain 75 percent of the
pensions initial purchasing power. Members contribute a percentage of monthly base pay in
excess of $513. The rate was 5 percent until November 2010, and is 8 percent thereafter.
Retiree Health. A member who starts a CalPERS pension within four months of leaving
service can continue participating in the medical and dental plans covering active employees;
after age 65, medical coverage is provided under a Medicare supplement plan. Coverage is
available for member, spouse and certain other eligible dependents.
Existing law defines a maximum state contribution tied to average costs for plans used by active
employees. For 2011, the state maximum contribution rate is $542 for one-party coverage, $1,030
for two-party coverage and $1,326 for other coverage. A retiree is eligible for a state contribution
equal to a portion of this maximum amount. For current employees, this portion is 0 percent with
less than ten years of service, and increases in steps from 50 percent at ten years until it reaches
100 percent with 20 years of service.
!

A retirees premium equals the difference between the state-determined cost of the health plan
providing the retirees coverage and the state contribution that the retiree qualifies for.
The state-determined health plan cost rates are based on the pooled experience of active and
retired members. Because actual retiree costs are generally higher than this pooled rate, there is
an implicit subsidy from the state that is over and above the explicit state contribution.
For members 65 and older, the state contribution is first used to cover the cost of the Medicare
supplement and dental coverage; after covering these payments, any excess is used to offset the
members Medicare Part B premium.
CalPERS Highway Patrol Employees
CHP employees receive an enhanced pension benefit, but do not participate in Social Security as
a result of their employment. The benefit factor is 3 percent beginning at age 50 except that
for those first hired after 2010, it is 3 percent reduced by 0.12 percent for each year payments
begin before age 55. The maximum benefit is 90 percent of average pay. The average is taken
over 12 months for those first hired before 2011, and 36 months for others. Effective July 1, 2010,
members contribute 10 percent of pay; significantly smaller contribution rates applied before
then. Other provisions mirror those for State non-safety employees, except that the free survivor
pension continuation is 50 percent rather than 25 percent.
The retiree health provisions applicable to CHP are similar to the provisions applicable to State
non-safety employees.
Public School Teachers: CalSTRS and School District
Pension. Members of this system do not participate in Social Security as a result of their
employment. After termination, a member with five years of service can start to draw a pension as
early as age 55, or age 50 with 30 years of service. The monthly amount equals the product of
highest average monthly pay rate, years of service and a benefit factor based on the age payments
begin. An additional longevity bonus of between $200 and $400 is available to members with at
least 30 years of service before 2011.
Pay is averaged over 12 months for those with at least 25 years of service, and over 36 months for
others. The benefit factor increases from 1.1 percent at age 50 to 2.4 percent at age 63 or older;
0.2 percent is added for those with at least 30 years of service, with the result limited to
2.4 percent. A $6,163 lump sum is paid upon the members death. In contrast to CalPERS,
where the employer pays the cost of an automatic 25 percent pension continuation to a survivor
(50 percent for CHP survivors), CalSTRS members bear the cost of any survivor continuation
through a reduced monthly benefit.
The benefit is increased by 2 percent (not compounded) annually regardless of inflation, and by a
supplemental payment needed to maintain 85 percent of the pensions initial purchasing power.
Members contribute 8 percent of pay (6 percent from 2001 through 2010).
Retiree Health. Benefits vary significantly among school districts. For our analysis, we assume
that benefits mirror those provided to state non-safety employees, except that we assumed that no
subsidy applies after age 65 (this benefit is relatively uncommon in school districts), and that the
maximum school district contribution is equal to two-thirds of the maximum state contribution
for coverage before age 65. The latter assumption is intended to reflect the wide range of pre- age
65 benefit levels offered by districts.
Local Agency Contracting With CalPERS:
Miscellaneous (Non-Safety) Employees
Retirement plans provided by local governments in California vary from entity to entity. They
can differ sharply in terms of benefit formulas, what is included in pensionable compensation,
and in their retiree health benefits. Many contract with CalPERS to provide pensions based on
design choices authorized by statute.


!
To illustrate the impact of the CFFR reform proposals on a more generous program than we
have so far considered, for the local non-safety system we model a pension design that includes
some (but not all) of the optional enhancements available to a contracting agency. The benefit
factor is equal to 2.5 percent less 0.1 percent for each year that payments begin before age 55. Pay
is averaged over 12 months, and includes certain items of special compensation. Members
contribute 8 percent of monthly base pay, except to the extent that employers have agreed to pick
up some or all of the employees share. Other features parallel those that apply to state members,
except that cost of living increases protect 80 percent of initial purchasing power (rather than
75 percent).
We assume that retiree health benefits mirror those provided to state employees, except that the
maximum employer contribution equals three-fourths of the maximum state contribution. This
reflects our observation that local retiree health plans vary a great deal and tend, on average, to
be less generous than the state.
Federal Employees System
The benefits described here apply to Federal employees first hired after 1983. These employees
participate in Social Security as a result of their employment.
Under this system, a member with five years of service can start to draw a pension as early as age
62, or, with ten years of service, as early as an age between 55 and 57, depending on year of birth.
The monthly amount is the product of highest average monthly pay rate, years of service, a
benefit factor and an early payment factor. Pay is averaged over 36 months. The benefit factor is
1.0 percent, except that it becomes 1.1 percent if the employee terminates at age 62 or older with
20 years of service. There are a variety of provisions related to early retirement, including a
temporary supplement payable until age 62 if certain conditions are met, and special provisions if
termination is in connection with a reduction in force or agency reorganization. There is an
annual cost-of-living increase after age 62 equal to the lesser of price inflation and 2 percent
where inflation is 3 percent or less, and inflation less 1 percent otherwise. Members contribute
0.8 percent of base pay.
Law enforcement members, who must generally retire by age 57, are subject to a variety of
special provisions. Their contribution rate is 1.3 percent of base pay. If they have sufficient years
of service (25, or 20 if retiring at age 50 or older) they receive significantly enhanced benefits
compared to other federal employees.
Federal employees also participate in a defined contribution plan. The employer matches
employee contributions up to 3 percent of base pay on a dollar-per-dollar basis, and on a 50-cent-
per-dollar basis for contribution between 3 percent and 5 percent of base pay. There is an
additional employer contribution of 1 percent of pay on a non-matching basis.
Those who draw a pension immediately after terminating service and their dependents can
continue to participate in the health plans covering active employees. Retirees pay the same
premiums for this coverage as employees.
Private Sector Comparison Group
Results for this group reflect an un-weighted average of the benefits provided for the general
workforces of six large private sector employers based in California: Chevron; Cisco; McKesson;
Northrop Grumman; Qualcomm; and Safeway. As noted earlier, large employers tend to provide
more generous benefits than their smaller counterparts, and we believe this group, as a whole,
provides benefit levels that are consistent with larger employers generally. This group also reflects
private sector trends previously noted:
!

Movement away from defined benefit plans especially traditional pensions:
While all of the California public sector employees considered in this report continue to earn
benefits under traditional pension plans, among our private sector group this is the case only
for Chevron employees hired before 2008. Chevron employees hired after 2007, Safeway
employees, and Northrop Grumman employees hired before July 2008 accrue defined benefits
under hybrid designs that differ significantly from a traditional pension. Like traditional
pensions, hybrid plans provide a benefit based on a formula. However, the benefits accrue in a
generally age neutral pattern, whereas a key characteristic of a traditional pension plan is that
a large portion of the total value is earned in late career. Also, benefits under a hybrid plan
are generally paid as a lump sum. Thus, while investment risk under a hybrid design remains
with the employer, longevity and inflation risk are usually borne by the retiree.
Employees of Cisco, McKesson and Qualcomm, and Northrop Grumman employees hired
since June 2008 do not earn benefits under any sort of defined benefit design.
Increased reliance on defined contribution plans. Unlike the California public sector
employees considered in this report, employees in our private sector group earn some or all of
their employer-provided retirement income benefits under a defined contribution plan, such as
a 401 (k) arrangement.
10
These plans provide age-neutral accruals, and employees bear all
investment, longevity and inflation risk. Employees who use the plans to save for their own
retirement receive employer-matching contributions that can be a large as 8% of pay (in the
case of Chevron), though the maximum employer match for most companies in the sample is
closer to 5% of pay. Under the typical arrangement, the first dollars an employee contributes
are matched more generously than additional employee contributions. Northrop Grumman
employees receive an additional employer contribution of between 3% and 5% of pay on a
non-matching basis, as a pension replacement benefit.
Movement away from providing retiree health benefits: Only Chevron still provides
an ongoing program; even here, the employer obligation is limited to the portion of future
costs that reflect no more than 4 percent annual health inflation since retirement. Otherwise no
benefit is available to employees within the sample group (except for isolated employees who
met grandfathering age and service requirements as of a past threshold date, who may still
become eligible for coverage upon future retirement).
Information about the private sector programs modeled in this report was taken from a variety of
publicly available sources, including descriptions provided to participants, summaries attached to
annual information returns filed with the Department of Labor, and various disclosure documents
filed with the Securities and Exchange Commission. If detailed historical information was not
available, we treated provisions in effect at a subsequent time as also applying to prior periods.
CFFR Alternatives
Both CFFR reform proposals would apply to current as well as future employees, beginning on
the respective effective dates. Alternative A applies to all employees upon declaration of a fiscal
emergency, which for purposes of our modeling is assumed to be January 2013. Alternative B is
implemented with respect to service and earnings after mid-2012; for modeling purposes, we
treated this effective date as January 2012.
Pension benefits accrued up to these operative dates are preserved. However, retirement income
benefits earned after that date will generally accrue more slowly, and under Alternative A some
employer retiree health care subsidies are reduced.

10
The exception is Safeway, which relies exclusively on a hybrid defined benefit design. This design,
though, closely resembles benefits provided by a defined contribution plan.


!
As a result, the pension amounts in our detailed comparisons under Alternatives A and B are a
combination of the benefits earned under the existing system for service up to the reform effective
date and the new system thereafter.
Alternative A
Retirement Income Benefits. This alternative bases retirement income benefits for service
after the fiscal emergency trigger date on the program covering federal employees (described
above), with the following modifications.
Pay cap. Pensions for future service would reflect only the portion of base pay up to
75 percent of the Social Security wage base (as defined on the effective date of the measure).
This cap would currently be about $80,000. The employer would contribute 3 percent
(4 percent for safety members) of pay in excess of the cap to the employees defined
contribution account.
Higher employee contributions. Employees would be required to contribute a
percentage of pay that reflects one-half of the normal cost for pension accruals. For purposes
of our modeling, we assumed this would result in a 3 percent of pay employee contribution
(7.5 percent for CHP), compared with 0.8 percent under the federal pension program
(1.3 percent for law enforcement). This contribution would apply to total base pay, including
any portion in excess of 75 percent of the Social Security wage base.
Social Security replacement. For employees not covered by Social Security, including
teachers and CHP employees, an additional pension would be provided with a value
approximating that of the employer-provided portion of a Social Security benefit allocable to
the years of service in the new system.
Retiree Health Care. Health plan rates would be determined based on the experience of
retirees and their dependents only. Other things being equal, this change will increase premiums
for retirees and reduce them somewhat for employees. For purposes of determining retiree
premiums for coverage prior to age 65, the percentage of the maximum state contribution that
otherwise applies is reduced by 5 percent for each year that retirement precedes age 62.
Post-65 retiree health coverage will not be available for the members spouse or other dependents,
and coverage for the member will be contingent on the member having made additional
pre-retirement contributions.
Alternative B
Retirement Income Benefits. This alternative places a limit on the amount of employer
funding associated with retirement income benefits to be earned after June 30, 2012. It does not
mandate a particular design under which those benefits accrue. For purposes of this study, it is
assumed that no pension benefits are earned for service after June 30, 2012, except for employees
not covered by Social Security. For the period after June 30, 2012, employees would earn
employer matching contributions under a defined contribution arrangement equal to the
employees own contribution up to 6 percent of base pay (9 percent for CHP employees); the
account balance based on these employer contributions would be vested after five years of service.
For the period after June 30, 2012, teachers would earn additional pension benefits that
approximate the value of a Social Security benefit first payable at age 62, and CHP employees
would earn additional pension benefits that approximate the value of a benefit first payable at age
57 equal to the estimated Social Security benefit first payable at age 62 without early payment
discount. Employees would contribute half the normal cost associated with this benefit. We
interpreted Alternative B to require normal costs for this purpose to be based on the assumptions
that apply to private sector employers.
Retiree Health Care. This alternative leaves existing law provisions unchanged.
!

Results of Our Detailed Comparisons
This section presents the results of our detailed retirement benefit calculations. The results are
expressed in present value terms. As discussed in the accompanying box, the present value
calculations convert future streams of payments into todays (2011) dollars. It enables us to present
different types of benefits on a comparable basis.
The graphs that follow include stacked columns that can be measured via the scale on the left of
each chart. Suppose a scenario involves an employee who is now (say) age 42 and who will
terminate service at age 57 in 2026. If the top of a column labeled CA Private Sector
corresponds to $400,000 on the left axis, it can be read as follows:
As of the employees termination of service in 2026, the value of his total retirement benefits under the
composite private sector program considered in this study is $400,000 (in 2011 dollars), based on the
assumptions used here.
The value would be different using different assumptions about such factors as how future benefit
payments should be discounted to reflect the time value of money. The focus should be on the
relative values for different programs, rather than particular dollar amounts. We also note that
the age, years of service, and pay amounts shown in the charts are as of January 2011. The pay
amounts are assumed to increase over time consistent with the actuarial assumptions used by
CalPERS and/or CalSTRS.
Comparison #1 Public Systems and Private Sector
Here we compare benefits under three of our four public sector groups; results for state non-safety
employees are shown for two designs, depending on whether the hire date was prior to January
14, 2011. We also include benefits under the federal employee program and our composite
private sector program. Other than teachers, each employee group participates in Social Security.
This first set of comparisons excludes both state (CHP) and federal law enforcement employees,
who receive significantly higher pensions than non-safety employees. We excluded them because
the sample of private sector benefit programs are for general (non-safety) employees.
Comparisons involve a current or recent hire at age 27 with an initial annual salary of $45,000.
We consider retirement at age 52, age 57, and age 62. We also look at early termination at
age 35.
Note that the employer-provided pension is less generous for this newly hired state employee than
for past hires, including those considered later in this chapter. Pensions for a new employee are
based on 36-month pay averaging, vs. 12-month for those hired before 2007. The recent increase
in employee contribution rate from almost 5% of pay to almost 8% will apply for the entire
career of a new hire.
The charts below show that the California local employee generally fares better than his
counterparts. This advantage would be still larger if, as in some jurisdictions, this local
government agency picked up some or all of the employee contribution.
Figure 6A provides comparisons for an employee who retires at age 52. For the public sector
funds, it shows that total employer-provided value is largest for the employee of our local
California public sector employer, and smallest for the public school teacher. The private sector
system yields benefits that are less than the state and local funds, but similar to the federal and
state teacher systems.


!
Figure 6A
Early Retirement at Age 52

0el|red Corl(|oul|or
0el|red 8erel|l
3oc|a| 3ecu(|lv
*Value of Net Employer-Provided Benefit at Termination of Service (2011 $)
Rel|(ee lea|lrca(e
-
0|d T|e( 1 NeW T|e( 1
3lale Nor-3alelv
3croo|
Teacre(
Loca| Nor-
3alelv
Fede(a|
Nor-3alelv
CA P(|vale
3eclo(
1,200,000*
1,000,000
800,000
00,000
400,000
200,000
Recent Hire:
Age 27
Annual Base Pay of
$45,000
Termination at age of
52

Pensions under the CalPERS design used by the local employer include generous early retirement
benefits that are important in this scenario. This is in contrast to CalSTRS, which has a benefit
formula that does not include generous early retirement factors.
The figure also shows that, for an employee who is not highly paid and retires early, retiree health
care benefits can easily be as valuable as the pension benefit; this is the case for the workers
covered by the state and local pension systems. By comparison, members covered under the
CalSTRS and federal programs do not receive any retiree medical benefits. The teacher in this
example has less than 30 years of service so pension payments cannot begin until age 55 (three
years after termination), and under the assumed school district design, retiree health care benefits
are therefore not available. For the non-safety federal employee, retiree health care benefits are
generally not available for termination before age 57.
Figure 6B shows comparisons for employees retiring at age 57. Under this scenario, both federal
and teacher employees benefits are improved relative to the previous figure. The federal
employee here qualifies for retiree health care benefits as well as a supplemental pension payable
until age 62. The teacher qualifies for certain pension enhancements tied to reaching 30 years of
service, and becomes eligible for immediate pension payment upon reaching age 55, triggering
retiree health care eligibility under the assumed school district design. Note that the pension
advantage of the local over the state program begins to diminish as the employee remains in
service beyond his early 50s, where early retirement subsidies under the local program are
greatest. All public systems provide benefits that are above the private sector system.
!

Figure 6B
Early Retirement at Age 57

0el|red 8erel|l
3oc|a| 3ecu(|lv
Rel|(ee lea|lrca(e
-
0|d T|e( 1 NeW T|e( 1
3lale Nor-3alelv
3croo|
Teacre(
Loca| Nor-
3alelv
Fede(a|
Nor-3alelv
CA P(|vale
3eclo(
1,200,000*
1,000,000
800,000
00,000
400,000
200,000
Recent Hire:
Age 27
Annual Base Pay of
$45,000
57

Figure 6C shows comparisons for employees who retire at age 62. For retirement at older ages,
the teachers pension is roughly on a par with the pensions for the other California public sector
employees considered here. But the absence of Social Security and the fact that few school
districts provide subsidized retiree health care coverage after age 65 means that the teachers total
employer-funded retirement benefit is still significantly smaller than what the public sector
comparators provide. The generous early retirement provisions under the California state and
especially local public sector pension designs are no longer a significant factor for age 62
retirements. The federal pension includes a 10% increase for retirement at age 62 with 20 years of
service, which more than replaces the value of the temporary supplement that is no longer
available.
Figure 6C
Retirement at Age 62

0el|red 8erel|l
3oc|a| 3ecu(|lv
Rel|(ee lea|lrca(e
Recent Hire:
Age 27
Annual Base Pay of
$45,000
62
-
0|d T|e( 1 NeW T|e( 1
3lale Nor-3alelv
3croo|
Teacre(
Loca| Nor-
3alelv
Fede(a|
Nor-3alelv
CA P(|vale
3eclo(
1,200,000*
1,000,000
800,000
00,000
400,000
200,000


!
Figure 6D provides comparisons for employees who terminate service at age 35. In all cases, the
employee terminates service before eligibility for retiree health benefits. As noted elsewhere in this
chapter, the value of pension accruals is low at younger ages, and accelerates rapidly later in an
employees career; on the other hand, employees contribute the same percentage of pay at all
ages. For the state non-safety employee, the members own contributions between ages 27 and 35
fund almost all of the pension, leaving only a small employer-provided value; for the teacher,
those employee contributions fund more than the value of the earned pension, leaving no
employer-provided value (despite vested status). The exception is the local employee, who
benefits from the more generous formulas provided by his or her local agency.
Because their defined contribution accruals are age-neutral, the federal and private sector
programs compare favorably to the state and teacher programs in this scenario.
Figure 6D
Termination at Age 35

0el|red 8erel|l
3oc|a| 3ecu(|lv
Rel|(ee lea|lrca(e
-
0,000*
50,000
40,000
30,000
20,000
10,000
Recent Hire:
Age 27
Annual Base Pay of
$45,000
35
0|d T|e( 1 NeW T|e( 1
3lale Nor-3alelv
3croo|
Teacre(
Loca| Nor-
3alelv
Fede(a|
Nor-3alelv
CA P(|vale
3eclo(

Assumptions for Present Value Calculations
Results in this chapter show employer-provided retirement benefits in terms of their present (lump
sum) value at termination of service, expressed in todays dollars. Comparison via present value
puts amounts with different payment terms pensions, defined contribution balances and retiree
health benefits onto a common basis. It also captures the value of important pension features
like cost-of-living adjustments, survivor benefits and temporary supplements.
The Appendix to this chapter contains detailed information on the assumptions used to determine
present value. Some of the key assumptions that go into present value calculations are: the
discount rate; future price inflation (general); future price inflation (health care); retiree health
care cost adjustment; investment return; and salary increases.
Discount Rate. This rate discounts for the time value of money, for the period between
termination of employment and each future benefit or premium payment. A rate of 6 percent per
year was used. Use of a higher rate would reduce the value of pension and retiree health benefits
but leave defined contribution values unaffected, and so would reduce the value of total benefits
more sharply for California public sector employees than for others. Use of a lower rate would
have the opposite effect.
!

Actuarial valuations of the California public sector pension plans considered here, which serve to
determine current employer costs, use a higher rate: 7.75 percent per year. This reflects expected
long-term investment returns. Consistent with guidance from the Government Accounting
Standards Board, a lower rate is generally used in determining costs for Californias unfunded
retiree health benefits: 4.50 percent per year.
This chapter does not attempt to measure benefit cost. Ultimately, the cost of a pension or retiree
health benefit varies from one employer to another, depending on factors like each employers
prefunding policy and investment policy, and its luck and skill in executing those policies.
Rather, the assumptions here are used to assign a value to the benefit. A retirees pension can be
considered to have a current value that is independent of whether future investment results and
other events mean that it will ultimately turn out to have cost her employer 5 percent of her
career pay or, instead, 15 percent. This value can be approached by considering what she would
have to pay to purchase the annuity stream from a private insurer.
The discount rates used by highly-rated insurance companies to price annuities have been
6 percent or less for a number of years. In the current market (March 2011), the rates range from
4.00 percent to 4.50 percent.
Although use of a lower discount rate could be justified based on current conditions, 6 percent
was chosen as an appropriate long-term assumption.
Future Price Inflation (General). This assumption is used in modeling the value of future
pension and Social Security cost of living adjustments. It is also used to express benefit values as of
a future termination of service in current dollars. A rate of 3.25 percent per year was used in this
study. A 3.00 percent rate is currently used in actuarial valuations of the California public sector
pension plans considered in this chapter, and a 3.50 percent rate is assumed in determining costs
for the pension plan covering federal employees.
Future Price Inflation (Health Care). This is a key assumption in modeling the value of
retiree healthcare benefits. A rate of 5 percent per year was used. In determining current costs,
California assumes an inflation rate that is 9.00 percent for 2012 and gradually reduces to
4.50 percent for 2019 and beyond, except that 4.50 percent is used for all years for dental benefits
and Medicare Part B premiums. Use of the exact California assumptions would result in larger
retiree health care values.
Retiree Health Care Cost Adjustment. Results in this study reflect the assumption that per
capita health care costs for retiree coverage prior to age 65 will exceed the cost that the state
assigns to that coverage for member premium purposes by 20 percent. The state-developed health
care cost rates reflect blended claims data covering both employees and retirees. Experience
shows that average costs for a retiree group are higher than for a group that also includes a large
number of employees, due to the increase in health care costs associated with increasing age.
Information provided in connection with the actuarial valuation of the State of California Retiree
Health Benefits Program indicates that an adjustment larger than 20 percent is supported by
claims data for CalPERS members, and was used in generating those valuation results; see the
report on the June 30, 2010 valuation prepared by Gabriel Roeder Smith & Company, pages
59-60. Use of a larger adjustment in this study would further increase the value of early retiree
health care benefits presented here.


!

Investment Return. This assumption is used to accumulate the value of employer
contributions under defined contribution plans, and employee contributions under pension plans,
or where employee prefunding of retiree health benefits is relevant. A rate of 7.25 percent per
year was used.
In determining the amount paid upon a refund of accumulated contributions, the California
public sector pension plans considered in this chapter credit interest at rates lower than
7.25 percent. But the purpose of the assumption here is to capture the economic sacrifice
represented by the employee contribution, rather than potential refund amounts.
Salary Increases. The stated assumptions in the actuarial valuations of the respective
California public sector pension plans were used. This means, for example, that a non-safety state
or local employee, a CHP employee and a public school teacher with the same current salary
have assumed pay levels that differ from one another in past and future years. The assumed
increase for a year is generally a function of the employees age and service.
Comparison #2 CalPERS State Employees
Here we look at retirement benefits for six employees in the CalPERS state miscellaneous (non-
safety) pension system, in various career stages and with different income levels and retirement
ages. We compare their benefits to those received under the federal system, our private sector
group, and to CFFRs alternatives.
New member hired under 2 percent at age 55 formula
Figure 7A shows the comparisons for a new employee, aged 27, who was hired prior to
January, 2011, and thus is covered by the old 2 percent at age 55 formula. This employee
works for 30 years before retiring at age 57. The employee is subject to the new 8 percent of pay
contribution rate for his whole career, and the 36-month average pay rule that applies to those
hired after 2006.
Figure 7A
CalPERS State: Full Career Employee
2 percent at Age 55

0el|red 8erel|l
3oc|a| 3ecu(|lv
** Fiscal Emergency under Alternative A, January 1, 2013
Rel|(ee lea|lrca(e
-
Cu((erl
Ca|PER3
Fede(a|
3vsler
CA P(|vale
3eclo(
A|le(ral|ve
8
1,800,000*
1,00,000
800,000
00,000
400,000
200,000
As of 1/1/11:
Age 27
0 Years of Service
Annual Base Pay Rate
of $45,000
State MisceIIaneous
57
1,400,000
1,200,000
1,000,000
A|le(ral|ve
A

!

Value of benefits. For this employee, the value of the employer-funded portion of total
CalPERS retirement benefits at termination of service in 30 years will be slightly over
$1 million, expressed in 2011 dollars. About 11 percent reflects the value of one-half (the
employer-funded part) of the portion of the expected Social Security benefit attributable to his
or her service for the state, 51 percent reflects the employer-funded portion of the CalPERS
pension, and the remaining 38 percent reflects the value of state subsidies for lifetime retiree
health coverage.
Comparison to federal system and private sector. Federal program benefits are only
9 percent less than those under the state program for this scenario. Note, however, that if
termination occurs any earlier than age 57 this gap would be far wider: for example, federal
benefits are 53 percent less for termination at age 56; this reflects that certain important
pension rights under the federal plan are earned all at once upon reaching key age and service
thresholds. The total state program benefit is well more than twice as large as the private
sector value.
Impact of CFFR alternatives. Each of the CFFRs alternatives would reduce total values
under the state program by roughly 30 percent. But that would still leave the employee with a
90 percent advantage over our private sector comparator.
Other issues. Compared to the CalPERS state program, all of the systems in the
comparison group have more risk sharing between the employer and employee.
New member hired under 2 percent at age 60 formula
Figure 7B provides similar comparisons, except that this employee was hired after 1/14/2011, and
thus is subject to the less generous 2 percent at age 60 formula under the new tier 1 design.
Figure 7B
CalPERS State: Full Career Employee
New Tier 1

0el|red 8erel|l
3oc|a| 3ecu(|lv
Rel|(ee lea|lrca(e
Cu((erl
Ca|PER3
Fede(a|
3vsler
CA P(|vale
3eclo(
A|le(ral|ve
8
A|le(ral|ve
A
-
1,800,000*
1,00,000
800,000
00,000
400,000
200,000
1,400,000
1,200,000
1,000,000
As of 1/1/11:
Age 27
0 Years of Service
of $45,000
State MisceIIaneous
57
CalPers Benefit Structure:
New Tier 1


!
Value of benefits. The value of benefits under the state program is about 17 percent less
than for the same employee hired right before January 14, 2011.
11
For this scenario, the value
of the retiree health benefit is almost equal to the value of the pension.
Comparison to federal system and private sector. The new tier 1 system provides
modestly less benefits than the federal system. Although the CalPERS formula is more
generous than the federal system formula, that is largely offset by a higher state-level employee
contribution (8 percent of pay, versus less than 1 percent for the federal employee). State
program benefits are still more than double the private sector values.
Impact of CFFR Alternatives. The CFFR alternatives would reduce this employees
benefit by about 19 percent in the case of Alternative A and 15 percent in the case of
Alternative B.
Young member working 10 years
Figure 7C shows a member who starts at age 27 and works for 10 years before leaving state
service.
Figure 7C
CalPERS State: Partial Career Employee

0el|red 8erel|l
3oc|a| 3ecu(|lv
Rel|(ee lea|lrca(e
-
0,000
50,000
40,000
30,000
20,000
10,000
70,000
80,000*
Cu((erl
Ca|PER3
Fede(a|
3vsler
CA P(|vale
3eclo(
A|le(ral|ve
8
A|le(ral|ve
A
As of 1/1/11:
Age 27
0 Years of Service
of $45,000
State MisceIIaneous
37
CalPers Benefit Structure:
New Tier 1

11
Employees hired after January 14, 2011 have an option of a new tier 1, shown here, or a tier 2,
which has a less generous benefit formula but no employee contributions. For the employee in this example,
the present value of lifetime employer provided benefits under tier 2 would be modestly lower than under
new tier 1 shown in the chart.
!

Value of benefits. Given the importance of age under pension plans, the shortfall in pension
value for an employee who terminates service at a relatively young age like 37 is much greater
than the proportional reduction in service years. In addition, the employee is not old enough
to earn retiree health care benefits.

Recall that the value of a pension grows very slowly in early years and accelerates rapidly in
later years. This pattern is exaggerated in cases where employee contributions are significant.
This is the case under the new tier 1 system, where employees contribute 8 percent of pay.
This contribution exceeds the value of the entire pension earned during the employees first
few years of service. An employee that terminates at a younger age will have made significant
contributions to the system and missed out on rapid accumulation of benefits that occurs in the
late stages of a career. Under the tier 2 (not shown), this employee would have earned a larger
net pension value despite the less generous formula, as no contributions are required under tier
2.
Comparison to federal system and private sector. This employee would receive larger
total benefits under both the federal system and the average private sector plan. This reflects
the age-neutral accruals under defined contribution plans.
Impact of CFFR Alternatives. This employee would earn larger benefits under each
reform alternative, largely because of the defined contribution features in each plan.
Member in Mid Career
This scenario reflects application of the provisions of Alternative A as of the assumed Fiscal
Emergency date (January 1, 2013) and the provisions of Alternative B as of July 1, 2012 to an
employee already in mid-career at those times.
Pension benefits accrued prior to the effective date of the reforms are preserved, and will continue
to grow with wage increases. However, benefits earned after that date will accrue more slowly,
based on the Alternative A and Alternative B provisions. The pension components for these two
alternatives represent a blend of benefits accumulated under the CalPERS Non-Safety employee
formula up to the effective dates, and under the alternatives thereafter.
Alternative A health care reforms operate differently than the pension reforms. An employee
retiring right after the effective date of the fiscal emergency might pay larger premiums for retiree
coverage than if she had retired right before that effective date. Although it is not considered in
this report, imposing comparable premium increases for those already retired at the effective date
would eliminate this imbalance. As mentioned previously, legal issues raised by the reforms,
especially as they apply to current system members, are outside the scope of this study.
The results for the federal system and the California private sector systems remain for reference.
They assume that benefits are accrued under these systems for the employees full career.
Figure 7D shows the impact on a mid-career employee who is now age 42 with 15 years of service
and $75,000 annual income. This employee is assumed to retire at age 57 (the last 13 or 14 years
of which would be under the alternative retirement systems).


!
Figure 7D
CalPERS State: Mid Career Employee

0el|red 8erel|l
3oc|a| 3ecu(|lv
Rel|(ee lea|lrca(e
Cu((erl
Ca|PER3
Fede(a|
3vsler
CA P(|vale
3eclo(
A|le(ral|ve
8
A|le(ral|ve
A
-
1,800,000*
1,00,000
800,000
00,000
400,000
200,000
1,400,000
1,200,000
1,000,000
As of 1/1/11:
Age 42
15 Years of Service
of $75,000
State MisceIIaneous
57

Value of benefits. The total benefit value if current law provisions remain in place would
exceed the total federal system value by 19 percent, and would be well more than twice the
composite private sector value.
Effects of CFFR alternatives applied to future accruals. Application of Alternative A
for the final 13 years of the employees career would reduce his benefit value by about 24
percent. Application of Alternative B for the final 13.5 years would result in a somewhat
smaller reduction: Alternative B does not attempt to reform retiree health care benefits. The
result of these reforms in this case would leave the employee slightly below the federal system
value, but still close to double the private sector benefit.
Highly Compensated Member In Mid-Career
This example illustrates the impact of the wage cap included in Alternative A. The employee in
this example has the same characteristics as the previous one, except for being more highly
compensated, earning $150,000 per year. Relative results are not much changed. However,
compared with the lower-paid employee just considered, more of this employees total retirement
benefit is delivered via the defined contribution plan, and so Alternative A further reallocates risk
to those most able to bear it.
!

Figure 7E
CalPERS State: Mid-Career Employee
Highly Compensated

0el|red 8erel|l
3oc|a| 3ecu(|lv
Rel|(ee lea|lrca(e
Cu((erl
Ca|PER3
Fede(a|
3vsler
CA P(|vale
3eclo(
A|le(ral|ve
8
A|le(ral|ve
A
-
1,800,000*
1,00,000
800,000
00,000
400,000
200,000
1,400,000
1,200,000
1,000,000
As of 1/1/11:
Age 42
15 Years of Service
of $150,000
State MisceIIaneous
57

Member in Late Career
This final example for state employees shows the impact of the alternative proposals on an
employee late in his career. This worker is 55, has 28 years of service as of January 2011, and is
retiring in three years at age 58. Since the fiscal emergency is assumed to apply only to service
and earnings after January 1, 2013, only the final year of this individuals pension would be
subject to the reduced accrual.
Figure 7F
CalPERS State: Late Career Employee

0el|red 8erel|l
3oc|a| 3ecu(|lv
Rel|(ee lea|lrca(e
Cu((erl
Ca|PER3
Fede(a|
3vsler
CA P(|vale
3eclo(
A|le(ral|ve
8
A|le(ral|ve
A
-
1,800,000*
1,00,000
800,000
00,000
400,000
200,000
1,400,000
1,200,000
1,000,000
As of 1/1/11:
Age 55
28 Years of Service
Annual Base Pay of
$100,000
State MisceIIaneous
58

Value of benefits. The value of net employer-provided retirement benefits under current
law exceeds those provided under the federal system by 31 percent, and exceeds those
provided under the average private sector plan by 124 percent.


!
Impact of CFFR alternatives. Prospective application of Alternative A would reduce this
members employer provided retirement benefit by about 11 percent to just under $1.1
million. The impact on the cash (pension) benefit would be minor, but the employee would
lose some retiree health subsidy. The reduction in benefits under Alternative B would be small,
as retiree health benefits would be unaffected.
Comparison #3: CalPERS California Highway Patrol Members
This section compares retirement benefits for five representative California Highway Patrol
(CHP) members to the amounts that peace officers (including the FBI) would receive under the
federal pension system. We also compare benefits under existing law to those earned under the
CFFR proposed Alternative A (modified federal system with Social Security replacement) and
Alternative B (a defined contribution system with an up to 9-percent employer match and
significantly enhanced Social Security replacement). We do not include the private sector group
in these comparisons because of the unique circumstances and job requirements applying to
peace officers.
New MemberFull Career
Figure 8A shows the comparisons for a new employee, aged 27, who is hired under the new
3 percent at age 55 formula. This employee works for 26 years and retires at age 53.
Figure 8A
CalPERS CHP: Full Career Employee
Post 2010 Hire (3 percent at age 55)

0el|red 8erel|l
3oc|a| 3ecu(|lv
Rel|(ee lea|lrca(e
-
Cu((erl
Ca|PER3
Fede(a| LaW
Erlo(cererl
A|le(ral|ve
8
2,200,000*
1,800,000
800,000
00,000
400,000
200,000
1,400,000
1,200,000
1,000,000
A|le(ral|ve
A
As of 1/1/11:
Age 27
0 Years of Service
Annual Pay Rate of
$73,000
53
CaIifornia
Highway PatroI
Reduced Calpers Benefit
Structure for Post-2010 Hire
2,000,000
1,00,000

Current law benefits. CHP employer-provided benefits are substantial, even after the
reduction in pre-age 55 pensions for a new hire like this example, and even though the recent
significant increase in employee contribution rates applies throughout his career.
Comparison to federal system. The total CalPERS benefit in this scenario is about even
with the total benefit provided to a career law enforcement employee under the federal system.
Effect of CFFR alternatives. The benefits would be reduced by about 39 percent under
CFFRs Alternative A, and by about 32 percent under CFFR Alternative B. Under Alternative
A, most of the reduction results from changes in retiree health care cost-sharing. Under
Alternative B the reduction is entirely due to smaller retirement income accruals.
!

Other issues. Defined contribution benefits play a significant role in the Federal system, as
well as the two CFFR alternatives. As noted earlier, unlike pension plans, defined contribution
plans leave various risks with the employee.
Member In Mid Career
This scenario involves the impact of prospective application of Alternative A and Alternative B to
an existing employee. Figure 8B shows the effect for an age 42 CHP member who has 15 years of
service as of January 2011, earns $100,000 per year, and retirees at age 53.
Figure 8B
CalPERS CHP: Mid-Career Employee

0el|red 8erel|l
3oc|a| 3ecu(|lv
Rel|(ee lea|lrca(e
-
Cu((erl
Ca|PER3
Fede(a| LaW
Erlo(cererl
A|le(ral|ve
8
0el|red 8erel|l
3oc|a| 3ecu(|lv
2,200,000*
1,800,000
800,000
00,000
400,000
200,000
1,400,000
1,200,000
1,000,000
A|le(ral|ve
A
As of 1/1/11:
Age 42
15 Years of Service
Annual Pay Rate of
$100,000
53
CaIifornia
Highway PatroI
Reduced Calpers Benefit
Structure for Post-2010 Hire
2,000,000
1,00,000
Rel|(ee lea|lrca(e

Value of benefits. In this scenario, because the employee was hired before 2011, the full
3 percent factor applies at age 53 retirement. This employee also benefits from having paid
lower contributions during the first half of his career (employer pickup of some or all of
employee contributions was a common feature of bargaining agreements prior to 2011). As a
result of these factors, the employer-provided benefit received by this employee is significantly
higher than the federal counterpart.
Effect of prospective application of CFFR alternatives. If a fiscal emergency is
declared and retirement accruals after January 1, 2013 are determined under Alternative A,
the present value of lifetime retirement benefits would be reduced by about 25 percent. Under
prospective application of Alternative B, they would be reduced by 17 percent. Again, this
difference in impact relates to the retiree health care reforms included in Alternative A but not
Alternative B. The two reform measures would have similar impacts on retirement income
benefits (though greater risk would be borne by the employee under Alternative B, given that
more of the total benefit would derive from defined contribution amounts).
Highly Paid Mid-Career Employee
This scenario shows the impact of prospective application of Alternative A and Alternative B to a
highly compensated member in mid-career. Figure 8C shows the effect for a CHP member age
42 with 15 years of service in 2011, who plans to retire at age 53.


!
Figure 8C
CalPERS CHP: Highly Paid Mid-Career Employee

0el|red 8erel|l
3oc|a| 3ecu(|lv
Rel|(ee lea|lrca(e
-
Cu((erl
Ca|PER3
Fede(a| LaW
Erlo(cererl
A|le(ral|ve
8
2,200,000*
1,800,000
800,000
00,000
400,000
200,000
1,400,000
1,200,000
1,000,000
A|le(ral|ve
A
As of 1/1/11:
Age 42
15 Years of Service
Annual Pay Rate of
$140,000
53
CaIifornia
Highway PatroI
2,000,000
1,00,000

Value of benefits. Under current law, this CHP employee retires with total employer-
provided benefits that are over one-third more than his federal counterpart has earned.
Effect of CFFR alternatives. Prospective application of Alternative A would reduce this
members benefit by about 30 percent, due to the lower accruals under the modified federal
system, the operation of the wage cap, and a reduction in the state subsidy for retiree health
benefits. The reduction under Alternative B would be 25 percent, due to the lower accrual of
cash retirement benefits.
Member in Late Career
This scenario shows the impact of prospective application of Alternative A and Alternative B to
an existing employee nearing the end of his career. Figure 8D shows the effect for a CHP member
age 50 with 23 years of service in 2011, who retires in 2014 at age 53.
!

Figure 8D
CalPERS CHP: Late Career Employee

0el|red 8erel|l
3oc|a| 3ecu(|lv
Rel|(ee lea|lrca(e
-
Cu((erl
Ca|PER3
Fede(a| LaW
Erlo(cererl
A|le(ral|ve
8
2,200,000*
1,800,000
800,000
00,000
400,000
200,000
1,400,000
1,200,000
1,000,000
A|le(ral|ve
A
As of 1/1/11:
Age 50
23 Years of Service
Annual Pay Rate of
$120,000
53
CaIifornia
Highway PatroI
2,000,000
1,00,000

Value of benefits. Under current law, this CHP employee retires with total employer-
provided benefits that are about 40 percent more valuable than his federal counterpart has
earned.
Effect of CFFR alternatives. Prospective application of Alternative A would reduce this
members benefit by about 12 percent, mostly due to an immediate reduction in the state
subsidy for retiree health benefits. However, a portion is attributable to reduced retirement
income benefit accrual during his final year of service, given the wage cap under this reform.
The reduction under Alternative B would be just 3 percent, as there is no retiree health benefit
impact.
Member Separating Before 20 Years
Figure 8E shows comparisons for a mid-career employee who is now age 45 with 10 years of
service, has annual pay of $110,000, and who retires at age 50 with 15 years of service.


!
Figure 8E
CalPERS CHP: Mid Career Employee Retiring at Age 50

0el|red 8erel|l
3oc|a| 3ecu(|lv
Rel|(ee lea|lrca(e
-
Cu((erl
Ca|PER3
Fede(a| LaW
Erlo(cererl
A|le(ral|ve
8
2,200,000*
1,800,000
800,000
00,000
400,000
200,000
1,400,000
1,200,000
1,000,000
A|le(ral|ve
A
As of 1/1/11:
Age 45
10 Years of Service
Annual Pay Rate of
$110,000
50
CaIifornia
Highway PatroI
2,000,000
1,00,000

Value of benefits. The comparison with federal system benefits becomes dramatic in this
scenario. This employee would need to have 20 years of service to qualify for the very
significant law enforcement pension enhancements under the federal system (absent
termination due to a layoff or similar event), and so receives only the modest benefits pictured
here. The CalPERS plan does not have this sort of eligibility cliff.
Impact of CFFR alternatives. This member would experience a significant 30-percent
benefit decline under Alternative A. Because he does not satisfy the 20-year requirement, his
pension accruals under Alternative A are sharply reduced from the CalPERS rates. Changes
to retiree medical further impact this employee. Alternative B would reduce the members
benefits by a more modest 13 percent, given the lack of retiree health benefit impact.
Caveat. An employee faced with these circumstances would be highly unlikely to retire
voluntarily at age fifty. However the alternative does illustrate the cliff effect in the federal
pension program, which is included in Alternative A for accruals after the triggering date.
Such an effect would strongly encourage peace officers to remain employed until the eligibility
threshold is met (20 years in this case).
Comparison #4: CalSTRS (Teachers)
This section considers current program retirement benefits for three public school teachers and
compares them with those available to federal and private sector employees, and models the
impact of the reforms under alternatives A and B. For purposes of these examples, we assume the
member is employed by a school district that provides pre-age 65 retiree health benefits with
about two-thirds of the explicit employer subsidy that the state provides, and no subsidy after age
65. As applied to teachers, Alternative A is assumed to have no impact on retiree health benefits.
New MemberFull Career
Figure 9A is an example for a new hire, aged 27, who earns $45,000 and retires at age 57.
!

Figure 9A
CalSTRS: Full Career Employee

0el|red 8erel|l
3oc|a| 3ecu(|lv
Rel|(ee lea|lrca(e
-
Ca|3TR3/
0|sl(|cls
Fede(a|
3vsler
CA P(|vale
3eclo(
A|le(ral|ve
8
1,00,000*
800,000
00,000
400,000
200,000
As of 1/1/11:
Age 24
0 Years of Service
of $45,000
57
Teacher
1,400,000
1,200,000
1,000,000
A|le(ral|ve
A

Value of benefits. Unlike the state employee previously considered, this teacher would
receive employer provided retirement benefits that are considerably less than would be
provided under the federal system, and only marginally greater than private sector benefits.
Effect of CFFR alternatives. The member would receive appreciably larger benefits under
Alternative A, which is to be expected given that the federal system provides greater values
than the teachers current program. There is a smaller increase under Alternative B.
New MemberFull Career In District With Higher Subsidies
School district policies can have a significant impact on the net amount of retirement benefits
received by teachers. The employee shown in Figure 9B is identical to the previous one, but is in a
district that (1) picks up a half of the required 8-percent employee pension contribution (employer
pickup of member contributions has been authorized since 2003, but our understanding is that it
is not common) and (2) provides retiree health care subsidies at similar levels to the State (up to
age 65). These district policies increase the value of current program benefits by about 40 percent.


!
Figure 9B
CalSTRS: Full Career Employee in District With
Higher Subsidies

0el|red 8erel|l
3oc|a| 3ecu(|lv
Rel|(ee lea|lrca(e
-
Ca|3TR3/
0|sl(|cls
Fede(a|
3vsler
CA P(|vale
3eclo(
A|le(ral|ve
8
1,00,000*
800,000
00,000
400,000
200,000
1,400,000
1,200,000
1,000,000
A|le(ral|ve
A
As of 1/1/11:
Age 24
0 Years of Service
of $45,000
57
Teacher
After 2002 District Funds
50% of Member's CalSTRS
Contribution
Vax|rur 3croo| 0|sl(|cl
Corl(|oul|or lo( Rel|(ee lea|lr Ca(e
P(|o( lo Ade 5: 100 ol Vax|rur
3lale Corl(|oul|or

Mid Career Member
Figure 9C looks at the prospective application of the CFFR alternatives to a teacher that is in
mid-career: now 42 years old with 18 years of service and annual income of $65,000, retiring at
age 57.
Figure 9C
CalSTRS: Mid Career Employee

0el|red 8erel|l
3oc|a| 3ecu(|lv
Rel|(ee lea|lrca(e
-
Ca|3TR3/
0|sl(|cls
Fede(a|
3vsler
CA P(|vale
3eclo(
A|le(ral|ve
8
0el|red 8erel|l
3oc|a| 3ecu(|lv
1,00,000*
800,000
00,000
400,000
200,000
1,400,000
1,200,000
1,000,000
A|le(ral|ve
A
As of 1/1/11:
Age 42
18 Years of Service
Annual Pay Rate of
$65,000
57
Teacher
Rel|(ee lea|lrca(e

This scenario demonstrates that a mid-career teacher is not significantly impacted by the reforms
considered here, though they face modestly higher investment risk.
!

Comparison # 5 Local Government
In this set of comparisons, we look at the effects of alternatives on four representative non-safety
members employed by a local public sector employer in California (agency). The benefit
provisions used for modeling purposes were summarized earlier, and are further described in the
Appendix to this chapter. Note that the pension formula recognizes certain elements of special
compensation in the employees final year; we assume that the includable special compensation
items boost final pay by 5 percent.
Some local employers agree to make some or all of the required employee pension contribution
(8 percent of pay for the agency considered here), as a means of increasing the employees total
compensation. We measure the potential impact of this feature in what follows.
New Member Working Full Career
Figure 10A shows the comparisons for a current hire, aged 27, now earning $45,000 annually.
This employee works for 30 years before retiring at age 57. In this example, the employee funds
the required pension contribution from his stated salary.
Figure 10A
CalPERS: Local Contracting Agency
Full Career Employee

Rel|(ee lea|lrca(e
0el|red 8erel|l
3oc|a| 3ecu(|lv
-
Ca|PER3
Erp|ove(
Fede(a|
3vsler
CA P(|vale
3eclo(
A|le(ral|ve
8
1,00,000*
800,000
00,000
400,000
200,000
1,400,000
1,200,000
1,000,000
A|le(ral|ve
A
As of 1/1/11:
Age 27
0 Years of Service
Annual Pay Rate of
$45,000
57
LocaI
Special Compensation in Final
Year Equals 5% of Base Pay

Value of benefits. Values for the current program are clearly larger than those for the
federal system and private sector comparators. As we saw in Figure 6B, the value also
surpasses benefits earned by the parallel state employee (even if under the richer 2 percent at
55 old tier 1 structure).
Effect of the CFFR alternatives. The CFFR alternatives would each reduce total benefits
for this employee by about 45 percent. Both cash (retirement income) and retiree health
benefits would be reduced under Alternative A, while all of the reduction in Alternative B
would be from reduced cash benefits. (Since alternative B takes effect shortly after this member
is hired, virtually all of his cash benefits are earned under the defined contribution plan.)
New Member In Local Agency With Employer Pickup
Some local employers agree to make some or all of the required employee pension contribution
(8 percent of pay for the agency considered here), as a means of increasing the employees total
compensation. This can have a substantial effect on the value of the employer-provided portion of
the pension benefit. The employee in Figure 10B is identical in all respects to the employee in the


!
previous example, except that his local agency funds his entire 8 percent of pay contribution in
addition to his stated pay level. (It is assumed that this employer funding is not itself treated as
additional retirement-eligible compensation.)
Figure 10B
Full Career Employee, Employer Picks Up Contributions

Rel|(ee lea|lrca(e
0el|red 8erel|l
3oc|a| 3ecu(|lv
-
Ca|PER3
Erp|ove(
Fede(a|
3vsler
CA P(|vale
3eclo(
A|le(ral|ve
8
1,00,000*
800,000
00,000
400,000
200,000
1,400,000
1,200,000
1,000,000
A|le(ral|ve
A
As of 1/1/11:
Age 27
0 Years of Service
Annual Pay Rate of
$45,000
57
LocaI
Employer Funds 100% of
Member's CalPERS
Contribution Since Hire
Rel|(ee lea|lrca(e
0el|red 8erel|l
3oc|a| 3ecu(|lv

The impact of the employer pick-up in this scenario is a 40 percent increase in the employer-
provided pension value, and a 25 percent increase in the total employer-provided value for all
benefits combined. The CFFR alternatives would have proportionally greater impacts on this
employee, reducing total retirement benefits by over one-half. These reforms do not
accommodate employer agreements to pick up a members required pension contribution.
Mid Career Member
Figure 10C shows the impact on a mid-career employee who is currently age 42 with 15 years of
service and who now earns $75,000 annually. This employee is assumed to retire at age 57 a
little less than half of his career service would take place after the effective dates of the reforms
considered here.
!

Figure 10C
Mid-Career Employee

Rel|(ee lea|lrca(e
0el|red 8erel|l
3oc|a| 3ecu(|lv
-
Ca|PER3
Erp|ove(
Fede(a|
3vsler
CA P(|vale
3eclo(
A|le(ral|ve
8
1,00,000*
800,000
00,000
400,000
200,000
1,400,000
1,200,000
1,000,000
A|le(ral|ve
A
As of 1/1/11:
Age 42
15 Years of Service
Annual Pay Rate of
$75,000
55
LocaI

Value of benefits. In this scenario, current program benefits exceed federal system (as well
as private sector) values by a dramatic margin. Here, termination prior to age 57 (absent a
layoff, major reorganization or similar event) would mean that the employee would not be
entitled to unreduced early payment of his basic pension, the temporary pension supplement
or retiree health care benefits under the federal system; so it is unlikely that the federal
employee in this scenario would actually terminate two years before age 57. A key message of
this scenario is that the federal program provides an employee with less retirement flexibility
than the CalPERS design.
Effects of CFFR alternatives. Application of Alternative A for the final 13 years of the
employees career would reduce his benefit by about 40 percent relative to the current system,
reflecting both the diminished value of pension accruals based on the federal program (for the
13 years after a Fiscal Emergency is effective) for termination before age 57, and reductions in
employer retiree health care subsidies. The reduction under prospective application of
Alternative B would be 21 percent. Nevertheless, this employee still fares better under the
reform alternatives than under the full federal program or the composite private sector
program.


!
Illustration of Modest Pension Spiking
There have been numerous accounts in the news media during recent years of extreme cases of
pension spiking at the local level, involving dramatic increases in final compensation due to late-
career pay increases and/or conversions of unused paid-leave entitlements. While the extreme
cases have been widely reported, it is important to note that even modest increases in final year
compensation can have a major impact on the lifetime value of a members pension.
Consider a 55 year old employee retiring after a 30-year career under the local agency plan just
considered (assuming no employer pick-up of member contributions). Suppose that her final
average pay reflects her final years base salary rate of $100,000. Her initial pension under the
Unmodified Allowance option is then $6,232/month. Taking into account the time value of
money (discount rate), life expectancy, future cost of living adjustments, survivor benefits and
other factors, this lifetime pension has a present value of about $1,333,000. Her own
contributions, accumulated with 7.25 percent interest, fund $376,000 of this total. So the net
employer-provided pension value is $956,000.
What if she is able to boost her final years pay for pension purposes by 10 percent? Her initial
monthly pension is then 10 percent larger at $6,865/month, and the lifetime value of the pension
also grows by 10 percent, to $1,468,000. But there is no change to her past contributions: the
accumulated value is still approximately $376,000. The net employer-provided pension value is
now $1,092,000 an increase of over 14 percent.
Looked at another way, the $10,000 of actual or imputed earnings that are used to boost her final
pay for pension purposes yields $136,000 in additional pension value.
Conclusion
The numerous scenarios in this chapter show the large disparity in pension benefits received by state
and local employees relative to private sector funds, and in most cases relative to federal law with
respect to early retirees. The one notable exception concerns teachers, where benefits are only
modestly higher than the private sector, and significantly below other California public sector
employees. The results also demonstrate the significant value of retiree health care subsidies.
Adoption of proposed alternative A would result in benefits levels that more closely align with the
federal system, though it is modestly less generous due to higher employee contribution rates and
the cap on pensionable compensation. Compared with benefits if the current programs continue
without change, this reform results in retirement benefit reductions in most scenarios we modeled.
The one exception is teachers, who would receive similar or larger benefits under the proposal.
Of course, no member would receive smaller pension benefits than already earned up to the
effective date of the reform.
Alternative B provides results similar to Alternative A in many of the scenarios. From a cash
(retirement income) benefit perspective, Alternative B is relatively more favorable to those leaving
at younger ages since (as modeled here) it relies mostly on defined contribution benefits for future
accruals. But the major difference between the impact of the two reforms stems from the fact
Alternative A includes important changes to retiree health programs, while Alternative B does not
address that benefit.
This chapter showed the effects of proposed pension changes on members of existing public sector
systems. Chapter 2 will look at the question of total compensation in the public sector versus private
sector, and Chapter 3 will model the effects of the CFFR alternatives on public sector employer costs.


Chapter 1: Appendix Tables 45

!
Chapter 1: Appendices
Benefit Provisions Reflected
Only provisions applicable to termination of service other than as a result of death or disability
are covered.
State Miscellaneous Employees
Defined Benefit Retirement Plans
CalPERS
These provisions apply to employees of the state of California who are members of SEIU Local
1000 and are covered by Social Security as a result of their employment. Unless electing Tier 2
coverage, Old Tier 1 applies to those hired before January 15, 2011 and New Tier 1 applies to
others.
Basic Pension Amount
A monthly pension beginning no earlier than the later of age 50 and the month following
termination of service is payable if the employee has at least five years of service and does not
receive a refund of accumulated contributions. The basic monthly amount is the product of years
of covered service, average pay, and a benefit factor.
Service includes 0.004 years for each day of unused sick or education leave at termination
of employment
Average pay is the average of the members full-time equivalent monthly pay rate during
the 12 consecutive months (36 consecutive months if first hired after 2006) over which the
average is highest, less $133.33
Benefit factor depends on age when payments begin:
!

46 Chapter 1: Appendix Tables


Age Old Tier 1 New Tier 1 Tier 2
50 1.100% 1.092% 0.50%
51 1.280% 1.156% 0.55%
52 1.460% 1.224% 0.60%
53 1.640% 1.296% 0.65%
54 1.820% 1.376% 0.70%
55 2.000% 1.460% 0.75%
56 2.064% 1.552% 0.80%
57 2.126% 1.650% 0.85%
58 2.188% 1.758% 0.90%
59 2.250% 1.874% 0.95%
60 2.314% 2.000% 1.00%
61 2.376% 2.134% 1.05%
62 2.438% 2.272% 1.10%
63 2.500% 2.418% 1.15%
64 2.500% 2.418% 1.20%
65 + 2.500% 2.418% 1.25%

Post-Retirement Death Benefits
The basic amount determined above is payable for the retirees lifetime, and 25% of that amount
continues for the remaining lifetime, if any, of the surviving spouse (or certain other statutory
beneficiaries). The retiree can elect to reduce the benefit so as to provide additional survivor
protection; conversion factors are significantly more generous for members first hired before
July 1, 1982.
In addition, a $2,000 lump sum is paid upon the retirees death.
Cost of Living Increases
Beginning the second calendar year following pension commencement, payments are
increased 2% annually on a compound basis provided that the cumulative increase
does not exceed cumulative price inflation since commencement.
An additional increase applies each year to the extent necessary to preserve 75% of the
pensions initial purchasing power provided that the total increase among all members
for a year does not exceed 1.1% of accumulated member contributions.
Member Contributions
Old Tier 1: prior to November 2, 2010, 5% of monthly base compensation in excess of
$513; subsequently, 8% of monthly base compensation in excess of $513
New Tier 1: 8% of monthly base compensation in excess of $513
Tier 2: Members do not contribute.


!
Contributions are accumulated with 6% annual interest and are returned where a pension benefit
is not payable.
"#$#%&'!()*'+,##-!.#/0%#)#1/!2,-/#)!3"(.24!
These provisions apply to federal employees first hired after 1983, other than firefighters, law
enforcement employees and members of certain other special groups.
Basic Pension
A monthly pension beginning no earlier than the later of the earliest commencement age and the
month following termination of service is payable if the employee has at least five years of service
and does not receive a refund of accumulated member contributions.
Earliest commencement age
A. if the employee has less than ten years of service, age 62
B. if (1) termination is in connection with a major reorganization, reduction in force
or transfer of function (special circumstances termination), and (2) either the
member has at least 20 years of service and is at least age 50 at termination, or
the member has at least 25 years of service regardless of age at termination, any
age
C. otherwise, the Minimum Retirement Age (MRA):
age 55 for members born before 1948
age 55 plus two months for each year that the member was born after 1947, for
members born after 1947 and before 1953
age 56 for members born after 1952 and before 1965
age 57 for members born after 1969
Prior to adjustment for form of payment, the monthly amount is the product of years of covered
service, average pay, a benefit factor and an early payment factor.
Covered service is rounded down to completed months; 50% of unused sick leave hours
at termination (100% for terminations after 2013) convert to additional service based on a
2087-hour year.
Average pay is an average of the members basic pay rate over the period of 36
consecutive months producing the highest such average.
The benefit factor is 1.0% except that it is instead 1.1% if the member is at least age
62 with at least 20 years of service at termination of employment.
Early payment factor
Factor is 100% if (i) payments begin on or after age 62, or (ii) payments begin on or after
age 60 and the member has at least 20 years of service, or (iii) the member has at least 30
years of service, or (iv) in the case of a special circumstances termination, payments begin
on or after age 55 and the member has at least 20 years of service.
Otherwise the factor is 100% less


1/6 of 1% for each month by which commencement precedes age 55 in the case of a
special circumstances termination, or
5/12 of 1% for each month by which commencement precedes age 62 in all other
cases.
Temporary Supplement
A supplement is payable in addition to the basic pension if the member
commences his basic pension with no early payment reduction, or
retires with at least 20 years of service after a special circumstances termination.
Payment of the supplement begins at the later of the time the basic pension commences and the
MRA. It ends at age 62, or upon the members death if earlier.
The supplement is a pro-rated portion of the estimated Social Security benefit earned as of
termination of service. For purposes of the estimate, Average Indexed Monthly Earnings are
determined as if
the period of included years equaled years from age 22 through termination of service,
less five
for each included year after hire and prior to termination, covered wages equaled the
members retirement eligible earnings (rather than actual FICA wages)
the member had no covered wages during or after the year of termination, and
for each year prior to hire and after age 22, if any, the member had covered wages that
progressed to assumed covered wages for the year of hire per past annual increases in
national average wages.
The estimated Social Security benefit is otherwise determined as if the member were age 62 at the
time supplemental payments begin. The pro-rated portion is 1/40 for each year of FERS service.
Payments can be forfeited if certain earnings limitations are exceeded.
The amount determined under the basic pension formula is the amount payable for the retirees
life only. In lieu of that the member can elect to receive a reduced basic pension:
90% of the formula amount during the retirees lifetime, with 50% of the formula amount
continued for the remaining lifetime, if any, of the designated beneficiary
continued for the remaining lifetime, if any, of the designated beneficiary.
No increase is provided prior to age 62. Subsequently the basic pension is increased annually by
the lesser of (i) the rate of price inflation and (ii) the greater of (1) 2% and (2) the rate of price
inflation less 1%.
Employee contributes a percentage of base pay equal to the excess of 7.00% over the OASDI
percentage (applicable to the employees base pay up to that years Social Security taxable wage


!
base) for the year. This employee contribution rate is 0.80% except as follows: 1.30% for 1987,
0.94% for 1988 and 1989, 1.05% for 1999, and 1.20% for 2000.
Contributions are accumulated with interest and returned to the member where a pension benefit
is not payable.
Alternative A
If a Financial Emergency (FE) is declared by the State, a members pension will generally be the
sum of a pension based on CalPERS provisions with respect to service prior to the FE, and a
pension based on FERS provisions with respect to service after the FE. The following
clarifications or exceptions apply:
1) The pension based on CalPERS provisions will reflect the members service as of the FE (as
if he or she terminated service on that date), the applicable CalPERS benefit factor based
on age at future pension commencement, and the larger of
average pay under the applicable CalPERS rules, determined as of the FE
average pay under FERS rules, determined as of termination of service.
The currently applicable CalPERS cost of living and post-retirement death benefit rules apply
to this benefit.
2) In determining the benefit based on FERS provisions, only service after the FE (as per
FERS rules) is used in computing the basic pension and in the pro-ration used to calculate
the temporary supplement.
The basic pension based on FERS provisions reflects a modified version of FERS average
pay. Under the modification, average pay is determined after limiting the members pay rate
for each month before or after the FE to 75% of 1/12 of the Social Security taxable wage
base (as determined under the law in effect as of March 31, 2011) for the year in which the
month falls.
The FERS cost of living and post-retirement death benefit rules apply to the resulting basic
pension.
3) Vesting in the pension based on CalPERS provisions is based on service through
termination of employment under CalPERS rules.
4) Vesting and other service-based eligibility requirements under the FERS provisions i.e.,
whether the member satisfies eligibility thresholds based on having 5, 10, 20, 25 or 30 years
of service are based on the members service both before and after the FE, determined
under FERS rules.
5) For the period after the FE, the employee contribution rate equals one-half of the normal
cost rate for the benefit in 2., as determined for contribution purposes with respect to this
group.
Alternative B
Alternative B places a limit on the amount of employer funding associated with retirement income
benefits to be earned after 2011. It does not mandate a particular design under which those
benefits accrue.
For purposes of this study, it is assumed that no pension benefits are earned for service after 2011.
A members pension would be based on:
service as of December 31, 2011, as if the employee terminated service on that date,


average pay under the applicable CalPERS provisions determined as of future
termination of service, and
the applicable CalPERS benefit factor based on age at future pension commencement.
Vesting in that pension is based on service through termination of employment under CalPERS
rules. The CalPERS cost of living and post-retirement death benefit rules continue to apply to
this benefit. Employee contributions discontinue after 2011.
Defined Contribution Retirement Plans
CalPERS
"#!$%&'#($)*+,-.$.!/$-$+01!
FERS
The Thrift Saving Plan (TSP) provides the following employer-funded benefits, subject to the
limits in Sections 402(g) and 415(c) of the Internal Revenue Code.
the account balance based on employer contributions of 1% of the members base pay. This benefit
vests after three years of service.
the account balance based on employer matching contributions. The employer matches an
employees contributions that are not in excess of 3% of base pay on a dollar-per-dollar
basis, and matches an employees contributions that are in excess of 3% of base pay but
not in excess of 5% of base pay on a 50-cent-per-dollar basis. Prior to June 2010,
matching contributions were not available prior to completion of one year of service.
This benefit is fully vested at all times.
Alternative A
For periods that follow the FE, members earn benefits on the same basis as TSP participants.
However, the account balance based on employer contributions of 1% of the members base pay
vests after five years of service (including service before and after the FE), rather than after three
years of service. In addition, the following benefit is provided:
The account balance based on employer contributions equal to 3% of the excess (if any)
of the members base salary for the month over 75% of 1/12 of the Social Security
taxable wage base for the year in which the month falls. This benefit vests after five years
of service (including service before and after the FE).
Alternative B
Members with five or more years of service (including service before and after July 1, 2012) are
vested in the account balance based on employer matching contributions. The match equals the
members own contributions made after June 30, 2012, up to 6% base salary.
Retiree Health Care Benefits
CalPERS
Eligibility: commence CalPERS pension within 120 days of separation from service
Coverage: retiree can elect coverage for self, spouse or certain other qualifying individuals; coverage
generally can continue for the individuals lifetime, provided that any required premiums are paid
Benefits: participation in any of the medical plans available to active employees prior to age 65,
and in a Medicare supplement plan thereafter; participation in any of the dental plans available to
active employees


!
Health Plan Rate: assigned monthly cost of participation for a year, determined by pooling
experience of active and retired populations
MSC: a monthly dollar amount determined by statute and updated annually; for 2011, the MSC
is $542 for one-party coverage, $1,030 for two-party coverage and $1,326 for other coverage
State %:
!
Date of First Hire State % (maximum is 100%)
Before 1985 100%
After 1984 but not after
January 1, 1989
10% for each year of service
After January 1, 1989
0% if less than ten years of service;
otherwise, 50% plus 5% for each year
of service in excess of ten years
Retiree Premium: the monthly premium for retiree or dependent for coverage equals
1) the excess, if any, of the total of the applicable Health Plan Rates over the product of the
State % and the applicable MSC
'$22!
2) the lesser of each covered individuals Medicare Part B premium and the excess, if any, of
the product of the State % and the applicable MSC over the total of the applicable Health
Plan Rates
FERS
Eligibility: commence FERS pension upon separation from service
Benefits: participation in any of the health plans available to active employees
Health Plan Rate: assigned monthly cost of participation for a year, determined by pooling
experience of active and retired populations
Retiree Premium: retirees are charged the same premium for participation in a given health plan as
active employees are charged
Alternative A
For those retiring after the FE, in determining retiree premiums for coverage prior to age 65:
1) The applicable Health Plan Rate reflects experience only for retirees and their covered
dependents.
!
2) The State % is the State % under current rules, reduced by subtracting 5% for each year
that pre-65 coverage begins prior to age 62.
For example, a member first hired in 2000 who retires with 18 years of service will have a State % of
90% under current rules; if retiring at age 55 after the FE, the State % with respect to premiums for pre-65
coverage would instead be 55% [= 90% 35% (= 5% x 7 years (= age 62 age 55))].
!


For those retiring after the FE, in addition to payment of retiree premiums for post-65 coverage,
such coverage will be available only for the member (and not for any of the members
dependents), and only if the member makes the required contribution for the minimum period
prior to retirement. The required contribution for a month is 50% of the average of the monthly
Health Care Rates among available Medicare Supplement plans for the year in which the month
falls. The minimum period is the lesser of ten years and the entire period from FE until
termination of service.
Alternative B
No provision
California Highway Patrol Employees
CalPERS
California Highway Patrol officers are not covered under Social Security as a result of their
employment.
termination of service is payable if the member has at least five years of service and does not
receive a refund of accumulated contributions. The basic monthly amount equals average pay
times the lesser of (i) 90% and (ii) the product of years of covered service and the benefit factor.
average pay is the average of the members full-time equivalent monthly pay rate during the
12 consecutive months (36 consecutive months if first hired after 2010) over which the
average is highest
service includes 0.004 years for each day of unused sick leave at termination of
employment
benefit factor is 3% reduced, if first hired after 2010, by 0.12% for each year by which
payments commence before age 55
The basic amount determined above is paid for the retirees lifetime, and 50% of it continues for
the remaining lifetime, if any, of the surviving spouse (or certain other statutory beneficiaries).
The retiree can elect to reduce the benefit to provide additional survivor protection. In addition,
a $2,000 lump sum is paid upon the retirees death.
pensions initial purchasing power provided that the total increase among all members
for a year does not exceed 1.1% of accumulated member contributions.


!
The member is assumed to contribute, or to have contributed in the past, a percentage of
monthly base compensation in excess of $863:
prior to July 1, 1995: 2.6%
from July 1, 1995 through June 30, 2007: 0.0%
after June 30, 2010: 10.0%
Contributions are accumulated with 6% annual interest and are returned where a pension benefit
is not payable.
Federal Employees Retirement System (FERS) for Law Enforcement Employees
These provisions apply to federal law enforcement employees first hired after 1983.
These employees participate in Social Security. They are generally subject to mandatory
retirement at the later of age 57 and completion of 20 years of service.
A qualifying law enforcement employee (QLEO) is at least age 50 with 20 years of service at
termination, or has at least 25 years of service regardless of age.
Basic Pension
A monthly pension beginning no earlier than the later of the earliest commencement age and the
month following termination of service is payable if the member has at least five years of service
and does not receive a refund of accumulated member contributions
Earliest commencement age
A. if the member has less than ten years of service, age 62
B. if the member is a QLEO, any age
C. otherwise, the Minimum Retirement Age (MRA):
age 55 for members born before 1948
age 56 for members born after 1952 and before 1965
age 57 for members born after 1969
Prior to adjustment for form of payment, the basic monthly amount is the product of average pay,
a benefit factor and an early payment factor.
!


Average pay is an average of the members basic pay rate over the period of
36 consecutive months producing the highest such average.
The benefit factor is 1% times years of covered service plus, if the member is a QLEO,
0.7% times 20 years of covered service (i.e., an additional 14%). Covered service is
rounded down to completed months; 50% of unused sick leave hours at termination
(100% for terminations after 2013) convert to additional service based on a 2087-hour
year.
Early payment factor:
The factor is 100% if the member is a QLEO.
Otherwise it is 100% less 5/12 of 1% for each month by which commencement
precedes age 62.
Temporary Supplement
A supplement is payable in addition to the basic pension if the member is a QLEO who
commences his basic pension immediately after termination.
Payment of the supplement begins immediately and ends at age 62, or upon the members death
if earlier.
The supplement is a pro-rated portion of the estimated Social Security benefit earned as of
termination of service. For purposes of the estimate, Average Indexed Monthly Earnings are
determined as if
the period of included years equaled years from age 22 through termination of service,
less five
members retirement eligible earnings (rather than actual
FICA wages)
the member has no covered wages during or after the year of termination, and
for each year prior to hire and after age 22, if any, the member had covered wages that
progressed to assumed covered wages for the year of hire per past annual increases in
national average wages.
The estimated Social Security benefit is determined as if the member were age 62 at termination.
The pro-rated portion is 1/40 for each year of FERS service. Payments can be forfeited if certain
earnings limitations are exceeded after the MRA.
The amount determined under the basic pension formula is the amount payable for the retirees
life only. In lieu of that the member can elect to receive a reduced basic pension:
continued for the remaining lifetime, if any, of the designated beneficiary
continued for the remaining lifetime, if any, of the designated beneficiary.


!
Post-Retirement Cost of Living Increases
If the member is a QLEO, increases apply both before and after age 62; otherwise they do not
apply prior to age 62. Increases apply to the basic pension only (not to the temporary
supplement). The annual increase is the lesser of (i) the rate of price inflation and (ii) the greater
of (1) 2% and (2) the rate of price inflation less 1%.
Employee contributes a percentage of base pay equal to the excess of 7.50% over the OASDI
percentage (applicable to the employees base pay up to that years Social Security taxable wage
base) for the year. This employee contribution rate is 1.30% except as follows: 1.80% for 1987,
1.44% for 1988 and 1989, 1.55% for 1999, and 1.70% for 2000.
Contributions are accumulated with interest and returned to the member where a pension benefit
is not payable.
Alternative A
sum of a pension based on CalPERS provisions with respect to service prior to the effective date
of the FE, and a pension based on FERS provisions with respect to service after that effective
date. The following clarifications or exceptions apply:
The pension based on CalPERS provisions will reflect the members service as of the FE
(as if he or she terminated service on that date), the applicable CalPERS benefit factor
based on age at future pension commencement, and the larger of
average pay under the applicable CalPERS rules, determined as of the FE effective
date
The currently applicable CalPERS cost of living and post-retirement death benefit rules apply to
this benefit.
In determining the benefit based on FERS provisions, only service after the FE effective
date (as per FERS rules) is used in computing the basic pension and in the pro-ration
used to calculate the temporary supplement. For purposes of the additional 0.7% benefit
factor per each of the first 20 covered years of benefit service applicable to a member
meeting the QLEO age and service requirements, only the excess, if any, of 20 years over
the members service as of the FE effective date under CalPERS provisions is reflected.
for each month before or after the FE effective date to 75% of 1/12 of the Social Security
taxable wage base (as determined under the law in effect as of March 31, 2011) for the year in
which the month falls.
pension.
Vesting in the pension based on CalPERS provisions is based on service through
termination of employment under CalPERS rules.
Vesting and other service-based eligibility requirements under the FERS provisions
i.e., whether the member satisfies eligibility thresholds based on having 5, 10, 20 or 25


years of service are based on the members service both before and after the FE,
determined under FERS rules.
For the period after the FE effective date, the employee contribution rate equals one-half
of the normal cost rate for the FERS-related benefit, as determined for contribution
purposes with respect to this group.
An additional pension benefit is provided whose value equals one-half (i.e., the employer-
funded portion) of the value of the Social Security benefit that the member would become
entitled to at age 62 or later termination if he or she had always been covered by Social
Security, to the extent attributable to service after the FE effective date. For purposes of
this study, this value was deemed to equal the value of the employer-funded Social
Security benefit earned by the Federal employee and attributable to his or her FERS
service, times the ratio of the CHP Officers post-FE effective date service to his or her
total service. It is expected that, as actually implemented, this benefit would be based on
alternative (simpler) provisions that provide a comparable value.
Alternative B
Alternative B generally places a limit on the amount of employer funding associated with
retirement income benefits to be earned after July 1, 2012, but does not mandate a particular
design under which those benefits accrue.
For purposes of this study, it is assumed that for an employee terminating service after
June 30, 2012, pension benefits equal the sum of A and B below.
A. A pension based on:
service as of June 30, 2012 based on CalPERS provisions, as if the employee had
terminated service on that date
average pay under the applicable CalPERS provisions determined as of future
the applicable CalPERS benefit factor based on age at future pension
commencement.
Vesting in this pension is based on service through termination of employment, determined
under CalPERS rules. The CalPERS cost of living and post-retirement death benefit rules
continue to apply to this benefit.
Employee contributions with respect to this benefit discontinue after June 30, 2012.
B. A pension whose value is comparable to the value of the following: a monthly
benefit equal to a pro-rated portion of the estimated Social Security Primary
Insurance Amount that the member would have earned as of termination of
employment with the State had he or she been covered by Social Security since
age 22, commencing at the later of age 57 and termination of service and payable
for the members life only, and subject to the same annual cost of living increase
as Social Security benefits then in payment status. The pro-rated portion equals
service after June 30, 2012 (as determined based on CalPERS provisions), limited
to no more than 35 years, divided by 35 years. The estimated Primary Insurance
Amount is determined as the amount first payable at the later of age 62 and
termination of service, without applying an early payment reduction and without
projecting changes to the bend points in effect for the year the member
terminates service. Average Indexed Monthly Earnings are determined as if


!
!
the period of included years equaled years from age 22 through the earlier of age 62
and termination of service, less five
members base earnings
the member has no covered wages during or after the year of termination with the
State, and
for each year prior to hire and after age 22, if any, the member had covered wages
that progressed to assumed covered wages for the year of hire per past annual
increases in national average wages.
Effective July 1, 2012, members contribute one-half of the normal cost associated with this
benefit, as determined under methods and assumptions consistent with those applicable to private
sector pension plans.
CalPERS
No employer-funded benefit
FERS
See the benefit summary for State Miscellaneous employees.
Alternative A
For periods that follow the FE, members earn benefits on the same basis as TSP participants.
However, the account balance based on employer contributions of 1% of the members base pay
vests after five years of service (including service before and after the FE), rather than after three
years of service. In addition, the following benefit is provided:
The account balance based on employer contributions equal to 4% of the excess (if any)
of the members base salary for the month over 75% of 1/12 of the Social Security
taxable wage base for the year in which the month falls. This benefit vests after five years
of service (including service before and after the FE).
Alternative B
Members with five or more years of service (including service before and after July 1, 2012) are
vested in the account balance based on employer matching contributions. The match equals the
members own contributions made after June 30, 2012, up to 9% base salary.
CalPERS
Eligibility: commence CalPERS pension within 120 days of separation from service
Benefits: participation in any of the medical plans available to active employees prior to age 65,
and in a Medicare supplement plan thereafter; participation in any of the dental plans available to
active employees
Health Plan Rate: assigned monthly cost of participation for a year, determined by CalPERS by
pooling experience of active and retired populations
!


MSC: a monthly dollar amount determined by statute and updated annually; for 2011, the MSC
is $542 for one-party coverage, $1,030 for two-party coverage and $1,326 for other coverage
State %:
!
Date of First Hire State % (maximum is 100%)
Before 1985 100%
After 1984 but not after
January 1, 1989
10% for each year of service
After January 1, 1989
0% if less than ten years of service;
otherwise, 50% plus 5% for each year
of service in excess of ten years
!
Retiree Premium: the monthly premium for retiree or dependent for coverage equals
1) the excess, if any, of the total of the applicable Health Plan Rates over the product of the
State % and the applicable MSC
'$22!
2) the lesser of each covered individuals Medicare Part B premium and the excess, if any, of
the product of the State % and the applicable MSC over the total of the applicable Health
Plan Rates
Employee Contributions: Beginning in July 2009, contributions to pre-fund retiree medical benefits
for California Highway Patrol officers began to be placed in an irrevocable trust. These
contributions were suspended in mid-2010, and are scheduled to resume in the future. Because
these contributions are either funded directly by the State, or via foregone salary increases (rather
than via reduction to the stated salary level used in the modeling undertaken here), the value of
employer-provided retiree health benefits was determined without regard to the projected value of
pre-funding contributions.
FERS
Alternative A
Alternative B
No provision.
Teachers
CalSTRS
These provisions apply to full-time teachers under the Defined Benefit Program. Their
employment does not give rise to participation in Social Security.


!
Service
Service is granted for the period for which the member makes contributions. Additional service is
granted equal to the ratio of unused sick leave days at termination of employment to the number
of days (excluding school and legal holidays) in the most recent school year. The member can
also purchase additional service periods. Primary eligibility service equals service, excluding
purchased service amounts and service derived from unused sick leave. Secondary eligibility
service equals primary eligibility service, plus up to 0.2 years derived from unused sick leave.
A monthly pension beginning no earlier than the later of age 55 (age 50 if the member has at least
30 years of service) and the month following termination of service is payable if the employee has
at least five years of primary eligibility service and does not receive a refund of accumulated
contributions. The basic monthly amount is (i) the product of years of service, average pay, and
an adjusted benefit factor, plus (ii) if the member had at least 30 years of secondary eligibility
service before 2011, a longevity bonus.
average pay. Average pay is the average of the members full-time equivalent monthly pay
rate during the period of 36 consecutive months over which the average is highest. The
period is instead 12 consecutive months if the member has at least 25 years of secondary
eligibility service. The period is also 12 consecutive months for certain classroom
teachers, if so provided under the applicable collective bargaining agreement; for
purposes of this study, no such bargaining provision is assumed.
adjusted benefit factor. The adjusted benefit factor is the lesser of (i) the factor from the
following table (based on age when payments begin) plus, if the member has at least
30 years of secondary eligibility service, 0.2%, and (ii) 2.4%.
!
Age Factor Age Factor
50 1.100% 57 1.640%
51 1.160% 58 1.760%
52 1.220% 59 1.880%
53 1.280% 60 2.000%
54 1.340% 61 2.133%
55 1.400% 62 2.267%
56 1.520% 63 + 2.400%
longevity bonus. The bonus is the lesser of (i) $200 plus $100 for each year by which total
secondary eligibility service exceeds of 30 years, and (ii) $400.
The monthly amount determined above is payable for the retirees lifetime only. The retiree can
elect to reduce the benefit so as to provide survivor protection. In addition to the pension, a
$6,163 lump sum is paid upon the retirees death.


Post-Retirement Increases
As of each September 1 following the first anniversary of pension commencement,
payments are increased by 2% of the initial pension amount, without regard to the rate of
price inflation, if any.
Each year an additional increase applies to the pension (other than the portion, if any,
based on a longevity bonus) to the extent necessary to preserve 85% of the initial
purchasing power provided that adequate funds are available within the State School
Lands Bank Fund and the Supplemental Benefit Maintenance Account.
Effective January 1, 2011, members contribute 8% of creditable compensation. They contributed
6% of creditable compensation during the period after 2000 and before 2011, and 8% of
creditable compensation prior to that. School districts may pay all or a portion of the
contribution on the members behalf, either as a device for reducing the portion of the members
compensation that is currently taxable, or (since 2003) as a means of also increasing the members
total (non-retirement eligible) compensation.
Contributions are accumulated with interest and are returned where a pension benefit is not
payable. Currently, the interest crediting rate for this purpose approximates the yield on two-year
Treasury notes.
Federal Employees Retirement System (FERS)
Alternative A
sum of a pension based on CalSTRS provisions with respect to service prior to the FE, and a
pension based on FERS provisions with respect to service after the FE. The following
clarifications or exceptions apply:
The pension based on CalSTRS provisions will reflect the members service as of the FE
(as if he or she terminated service on that date), the applicable CalSTRS adjusted benefit
factor based on age at future pension commencement, and the larger of
average pay under the applicable CalSTRS rules, determined as of the FE
The currently applicable CalSTRS post-retirement death benefit and post-retirement benefit
increase rules apply to this benefit.
In determining the benefit based on FERS provisions, only service after the FE (as per
FERS rules) is used in computing the basic pension and in the pro-ration used to calculate
the temporary supplement.
for each month before or after the FE to 75% of 1/12 of the Social Security taxable wage
base (as determined under the law in effect as of March 31, 2011) for the year in which the
month falls.
pension.


!
!
Vesting and other service-based eligibility requirements with respect to the benefit based
on CalSTRS provisions i.e., whether the member satisfies eligibility thresholds based
on having 5, 25 or 30 years of service are based on the members service both before
and after the FE, determined under CalSTRS rules.
Vesting and other service-based eligibility requirements with respect to the benefit based
on Fers provisions i.e., whether the member satisfies eligibility thresholds based on
having 5, 10, 20, 25 or 30 years of service are based on the members service both
before and after the FE, determined under FERS rules.
For the period after the FE, the employee contribution rate equals one-half of the normal
cost rate for the benefit based on Fers provisions, as determined for contribution purposes
with respect to this group.
An additional pension benefit is provided whose value equals one-half (i.e., the employer-
funded portion) of the value of the Social Security benefit that the member would become
entitled to at age 62 or later termination if he or she had always been covered by Social
Security, to the extent attributable to service after the FE effective date. For purposes of
this study, this value was deemed to equal the value of the employer-funded Social
Security benefit earned by the Federal employee and attributable to his or her FERS
service, times the ratio of the educators post-FE effective date service to his or her total
service. It is expected that, as actually implemented, this benefit would be based on
alternative (simpler) provisions that provide a comparable value.
Alternative B
Alternative B places a limit on the amount of employer funding associated with retirement income
benefits to be earned after July 1, 2012. It does not mandate a particular design under which
those benefits accrue.
For purposes of this study, it is assumed that for an employee terminating service after June 30,
2012, pension benefits equal the sum of A and B below.
A. A pension based on:
service as of June 30, 2012 based on CalSTRS provisions, as if the employee had
terminated service on that date
average pay under the applicable CalSTRS provisions determined as of future
the applicable CalSTRS benefit factor based on age at future pension
commencement.
Vesting in this pension is based on service through termination of employment, determined
under CalSTRS rules. The currently applicable CalSTRS post-retirement death benefit and
post-retirement benefit increase rules apply to this benefit. Employee contributions with
respect to this benefit discontinue after June 30, 2012.
B. A pension whose value is comparable to the value of the following: a monthly
benefit equal to a pro-rated portion of the estimated Social Security Primary
Insurance Amount that the member would have earned as of termination of
employment with the State had he or she been covered by Social Security since
age 22, commencing at the later of age 62 and termination of service and payable
for the members life only, and subject to the same annual cost of living increase


as Social Security benefits then in payment status. The pro-rated portion equals
service after June 30, 2012 (as determined based on CalPERS provisions), limited
to no more than 35 years, divided by 35 years. The estimated Primary Insurance
Amount is determined as the amount first payable at the later of age 62 and
termination of service, without projecting changes to the bend points in effect for
the year the member terminates service. Average Indexed Monthly Earnings are
determined as if
the period of included years equaled years from age 22 through the earlier of age 62
and termination of service, less five
members base earnings
the member has no covered wages during or after the year of termination with the
State, and
for each year prior to hire and after age 22, if any, the member had covered wages
that progressed to assumed covered wages for the year of hire per past annual
increases in national average wages.
Effective July 1, 2012, members contribute one-half of the normal cost associated with this
benefit, as determined under methods and assumptions consistent with those applicable to
private sector pension plans.
CalSTRS
FERS
Alternative A
Alternative B
CalSTRS
Except for payment of Medicare Part A premiums in certain cases, no benefit is provided by
CalSTRS. Employer-subsidized retiree health care benefits are provided by certain school
districts. The eligibility, cost-sharing and benefit provisions of these arrangements differ
significantly from district to district.
For purposes of this study it is assumed that the members district provides retiree health care
benefits prior to age 65 under eligibility and cost-sharing rules that are comparable to those
provided to State Miscellaneous employees, but with a maximum district contribution equal to
two-thirds (unless indicated otherwise) of the maximum state contribution, and with no employer-
subsidized benefits after age 65. See the benefit summary for State Miscellaneous employees for
information on pre-age 65 retiree health care benefits available to those employees.


!
FERS
Alternative A
No provision
Alternative B
No provision
Local Non-Safety Employees
Retirement plans provided by local governmental units in California counties, municipalities,
agencies vary from entity to entity. Some contract with CalPERS to provide retirement
benefits based on design choices within a limited menu of options authorized by statute. The
contracting-in design summarized here is neither the most nor the least generous available. It is
assumed to apply to miscellaneous (e.g., non-safety) employees who participate in Social Security
as a result of their employment. Only provisions applicable to termination of service other than
as a result of death or disability are covered.
CalPERS
termination of service is payable if the employee has at least five years of service and does not
receive a refund of accumulated contributions. The basic monthly amount is the product of years
of covered service, average pay, and a benefit factor.
service includes 0.004 years for each day of unused sick leave at termination of
employment
average pay is the average of the members full-time equivalent monthly pay rate and
certain items of special compensation during the 12 consecutive months over which the
average is highest, less $133.33
benefit factor is 2.5%, less 0.1% for each year payments begin before age 55
The basic amount determined above is payable for the retirees lifetime, and 25% of that amount
continues for the remaining lifetime, if any, of the surviving spouse (or certain other statutory
beneficiaries). The retiree can elect to reduce the benefit so as to provide additional survivor
protection.
Cost of Living Increase
pensions initial purchasing power.


Members contribute 8% of eligible compensation. Employers may pay all or a portion of the
contribution on the members behalf, either as a device for reducing the portion of the members
compensation that is currently taxable, or as a means of also increasing the members total (non-
retirement eligible) compensation. Contributions are accumulated with 6% annual interest and
are returned where a pension is not payable.
Federal Employees Retirement System (FERS)
Alternative A
Alternative B

CalSTRS
FERS
Alternative A
Alternative B
Current
Like other retirement benefits, the level of employer-subsidy for retiree health care varies
significantly among local governmental entities. For purposes of this study it is assumed that the
members employer provides retiree health care benefits under eligibility and cost-sharing rules
comparable to those provided for State Miscellaneous retirees, but with a maximum employer
contribution equal to 75% of the maximum state contribution. See the benefit summary for State
Miscellaneous employees for information on retiree health care benefits available to those
employees.
FERS
Alternative A
No provision
Alternative B
No provision

Chapter 1: Methods and Assumptions 65

!

Method and Assumptions
The results show the present value of net employer-provided retirement benefits as of future
termination of service, expressed in 2011 dollars. Accumulated employee contributions and
retiree premium payments toward funding the benefit are subtracted to arrive at the net
employer-provided value, and amounts separately contributed by the employee (for example, to
attract employer matching contributions) are not included. The net value as of future termination
of service is expressed in 2011 dollars by discounting for projected price inflation.
Sample Employee
!"#$%&#'()*'"+,-'#,".+*/)'/.0/,+&" the employees age, years of service and annual base pay rate
as of January 1, 2011, and age as of termination of service. Unless explicitly indicated otherwise,
it is assumed that termination is not in connection with a major reorganization, reduction in force
or transfer of function (special circumstances), increasing FERS-based benefits.
State Miscellaneous Employees
Unless indicated otherwise, the employee is assumed to have been hired
prior to January 15, 2011.
General
annual discount rate: 6%
mortality: for the period after payment commencement, the static healthy annuitant
mortality rates, by gender, mandated for use by large private sector employers in
determining required contributions to tax-qualified pension plans for years beginning in
2011, as per Internal Revenue Service Notice 2008-85; no mortality prior to
commencement
annual base pay increases: per the rates applicable to State Miscellaneous Tier 1 & Tier 2
members summarized on page A-4 of the report on the June 30, 2009 CalPERS State
and Schools Actuarial Valuation (rates not shown derived by linear
interpolation/extrapolation), with increases effective as of each January 1
retirement eligible non-base pay, where relevant: 10% of base pay
employee gender: male
future annual price inflation: 3.25%
continuity of service: participant entered plans at earliest eligibility after hire and was
employed in a covered position on a full-time basis until termination
value of employer contributions under defined contribution plans and employee
contributions under defined benefit plans: accumulated to termination with 7.25%/year
interest from assumed semi-monthly deposit
Alternative A: Fiscal Emergency as of January 1, 2013
Alternative B: treated as if effective January 1, 2012 rather than July 1, 2012
Pension
time of benefit payment or benefit commencement: the earliest eligible date following termination
of service (Chevron employee first hired before 2008: not earlier than age 50)
marital status: married to spouse of same age

66 Chapter 1: Methods and Assumptions

elected form of payment:
CalPERS: Option 2W, spouse is designated beneficiary
FERS-based: joint & 5/9 survivorship annuity, spouse is designated beneficiary
Chevron employee first hired before 2008: single life annuity
all others: lump sum
future annual increases in national average wages: 3.50%
annualized rate of return on 30-year Treasuries (Safeway): 4.25%
Social Security wages: 112% of base pay; for each year prior to hire and after age 22, if any,
employee is assumed to have covered wages that progressed to covered wages for year of
hire in keeping with past annual increases in national average wages; for each year after
termination and prior to age 62, if any, employee is assumed to have covered wages that
progress from covered wages for final year prior to termination in accordance with
assumed future annual increases in national average wages
Social Security benefit: the amount payable for the members life, assuming that the spouse is
entitled to an equal benefit based on his or her own covered wage history
employer-provided portion of Social Security benefit: 50% times the ratio (not in excess of 100%) of
years of service with the employer to 35 years
limit on CalPERS Purchasing Power Protection Allowance: the annual limit based on 1.1% of
accumulated member contributions is assumed not to apply
unused sick leave at termination of employment: three days per year of service
CalPERS program elections:
four years after enrolling in the Alternate Retirement Program, employee did or will
transfer Program funds to CalPERS
employees subject to Tier 2 provisions did or will convert to Tier 1
purchase of additional CalPERS service credits: none
gross normal cost rate for post-FE Alternative A accruals, for purposes of employees pay
half member contributions: 6% of base pay
earnings limitation with respect to FERS temporary supplement: assumed not to apply
"#$%&#'!()&*+%,-*%)&!.&'!/#*%+##!0#.1*2!3#&#$%*4!
matching contribution arrangements: employee always contributes the amount necessary to
attract the maximum employer matching contribution
pre-retirement withdrawals: none
retiree health benefits:
whenever eligible, medical, prescription drug and dental coverage for retiree will be
elected, and continue until death; value of retiree-only coverage increased by 65% to
reflect expected incidence of coverage of retirees spouse or other eligible dependents
employee will make contribution required for post-65 coverage during the initial
period following a Fiscal Emergency, for up to ten years (Alternative A)
after 2011, all measures tied to health care cost will increase by 5%/year
under CalPERS, the total of the applicable medical and dental plan rates will equal
100% of the MSC for coverage prior to age 65, and 67% of the MSC thereafter


!

the Federal retiree will elect the Blue Cross and Blue Shield Service Benefit Plan
(Standard) for pre-65 health coverage (rate = $578.61/month and member premium
= $187.18/month for one-party coverage in 2011), and the Blue Cross and Blue
Shield Service Benefit Plan (Basic) for post-65 health coverage (rate =
$453.48/month and member premium = $113.37/month for one-party coverage in
2011)
retiree claims cost will be 120% of the health plan rate where the coverage is primary
and the rate is based on blended active and retired population experience, and 100%
of the rate otherwise
Medicare Part B premium is the amount without increase due to income in excess of
threshold, or due to payment other than via Social Security withholding
California Highway Patrol Employees
General
mortality rates, by gender, mandated for use by large private sector employers in
determining required contributions to tax-qualified pension plans for plan years
beginning in 2011, as per Internal Revenue Service Notice 2008-85; no mortality prior to
commencement
annual base pay increases: per the rates applicable to CHP members summarized on page
A-5 of the report on the June 30, 2009 CalPERS State and Schools Actuarial Valuation
(rates not shown derived by linear interpolation/extrapolation), with increases effective
annually as of each January 1
value of employer contributions under defined contribution plans and employee contributions under defined
benefit plans: accumulated to termination with 7.0%/year interest from assumed semi-
monthly deposit
Alternative A: Fiscal Emergency as of January 1, 2013
Pension
of service
elected form of payment, where applicable: life annuity, with 50% of the amount payable during
members life (50% of the amount payable prior to reduction for joint & survivor form,
for FERS-based pension) continuing for surviving spouses remaining lifetime, if any


limit on CalPERS Purchasing Power Protection Allowance: the annual limit based on 1.1% of
accumulated member contributions is assumed not to apply
unused sick and education leave at termination of employment: three days per year of service
purchase of additional CalPERS service credits: none
gross normal cost rate for purposes of future employees pay half member contributions:
post-FE FERS-based accruals under Alternative A: 15% of base pay
post-2011 accruals under Alternative B (private sector assumptions): 11% of base pay
Defined Contribution and Retiree Health Benefits
employee will make contribution required for post-65 coverage during the initial
period following a Fiscal Emergency, for up to ten years (Alternative A)
under CalPERS, the total of the applicable medical and dental plan rates will equal
100% of the MSC for coverage prior to age 65, and 67% of the MSC thereafter
2011)
Medicare Part B premium is the amount without increase due to income in excess of
threshold, or due to payment other than via Social Security withholding


!

Teachers
Unless explicitly indicated otherwise, the following special conditions are assumed not to apply:
maximum school district contribution for pre-65 retiree health care is other than two-
thirds of maximum state contribution
a specified portion of the member contribution being funded by the employer and not
from the members stated pay rate (but not prior to 2003).
General
mortality rates, by gender, mandated for use by large private sector plans in determining
required employer contributions to tax-qualified pension plans for plan years beginning in
2011, as per Internal Revenue Service Notice 2008-85; no mortality prior to
commencement
annual base pay increases: 4%
retirement-eligible non-base pay, where relevant: 10% of base pay
Alternative A: Fiscal Emergency effective January 1, 2013
Pension
of service (Chevron employee first hired before 2008: not earlier than age 50)
CalSTRS, FERS-based and Chevron employee first hired before 2008: annuity form
providing largest payment during members life


Social Security wages: 112% of base pay; for each year prior to hire and after age 22, if
any, employee is assumed to have covered wages that progressed to covered wages for
year of hire in keeping with past annual increases in national average wages; for each year
after termination and prior to age 62, if any, employee is assumed to have covered wages
that progress from covered wages for final year prior to termination in accordance with
Social Security benefit: the amount payable for the members life, assuming that the
spouse is entitled to an equal benefit based on his or her own covered wage history
employer-provided portion of Social Security benefit: 50% times the ratio (not in excess
of 100%) of years of service with the employer to 35 years
purchase of additional service credits: none
unused sick leave at termination of employment: four days per year of service
days in school year, excluding holidays: 175 (for purposes of imputed service based on unused
sick leave)
gross normal cost rate for purposes of employees pay half member contributions:
for post-FE Alternative A accruals, 6% of base pay
for Alternative B (private sector assumptions), 4% of base pay
2011)
Local Miscellaneous Employees
Unless explicitly indicated otherwise, it is assumed that member contributions are funded from
the employees stated pay rate..
General


!

mortality: for the period after payment commencement, the static healthy annuitant mortality
rates, by gender, mandated for use by large private sector plans in determining required
employer contributions to tax-qualified pension plans for plan years beginning in 2011, as per
Internal Revenue Service Notice 2008-85; no mortality prior to commencement
annual base pay increases: same as for State Miscellaneous Employees
retirement-eligible non-base pay, where relevant: 10% of base pay
Alternative A: Fiscal Emergency effective January 1, 2013
Pension
of service (Chevron employee first hired before 2008: not earlier than age 50
CalPERS, FERS-based and Chevron employee first hired before 2008: annuity form
providing largest payment during members life
purchase of additional service credits: none
gross normal cost rate for post-FE Alternative A accruals, for purposes of employees pay half member
contributions: 6% of base pay


2011)

Chapter 2: Compensation 73

!
Chapter 2: Compensation
Introduction
In the first chapter of this report we found that pension and retiree health benefits received by
state and local government employees are considerably higher than those offered in the private
sector. Of course, retirement benefits are only part of the full compensation picture. It is also
necessary to consider wages and other benefits in order to make a valid comparison of public
sector and private sector compensation. This chapter looks at total compensation, focusing first
on wages then on benefits.
Background
As shown in Figure 1, about 70 percent of total compensation for all civilian workers is related to
wages and the remaining 30 percent is related to benefits.
!
The benefits include supplemental pay
(such as overtime premiums, bonuses, and stock options), paid leave, health insurance, retirement
benefits, and legally required benefits (such as social security, Medicare, and unemployment
insurance). The mix between wage and non-wage compensation is significantly different in the
public and private sectors. Non-wage benefits account for 34 percent of total compensation in the
state and local government sector, but only 29 percent in the private sector.
Figure 1
Major Components of Compensation
Civilian Employees, 2010

In the subsequent sections of this chapter, we examine the wage and non-wage components in
more detail. We first look at wage comparisons by analyzing occupational survey data and recent

1
Source: Bureau of Labor Statistics. Employer Costs For Employee Compensation (ECEC).
http://www.bls.gov/ncs/ect/

74 Chapter 2: Compensation

statistical studies. We then turn to a comparison of non-wage benefits offered in the public and
private sectors.
The appendix tables at the end of this chapter include detail on occupational wage comparisons
as well as summaries of recent studies, surveys, and other resources related to public versus private
compensation.
Comparison of Wages
As shown in Figure 2, average hourly wages for all state and government employees exceeds that
of private sector employees by 53 percent in the Los Angeles Combined Statistical Area (CSA),
22 percent in the San Francisco-San Jose-Alameda CSA and 35 percent in the greater
Sacramento CSA.
2
However, a more detailed analysis of the data reveals that aggregate wage
comparisons provide a misleading picture of comparative wage levels for specific jobs. There is a
substantial difference in the composition and level of occupations between the two sectors. In
general, workers in the state and local government sector have more education, are more
Figure 2
Average Hourly Earnings All Occupations
State and Local Government versus Private Sector, 2010

experienced, and are employed in higher skilled jobs when compared to the private sector as a
whole. Management, professional, and administrative support occupations account for two-thirds
of the state and local government workforce, compared with two-fifths of private industry.
3
In

2
Source: National Compensation Survey: Bureau of Labor Statistics. http://www.bls.gov/eci/. A
description of the areas covered by the CSAs is provided in the introduction to Appendix Tables 1
through 3.
3
Source: Employer Costs For Employee Compensation. News Release, Technical Note. BLS.
http://www.bls.gov/news.release/ecec.tn.htm


!
2009, about 55 percent of Californias state and local government employees had a college
degree, compared to about 35 percent of the private sector workers.
4

Analysis of BLS National Compensation Survey Data
To account for the differences in skill levels and occupational mix, it is necessary to look at wage
comparisons for similar jobs. To do this, we analyzed detailed occupational data from the most
current National Compensation Surveys (NCS) conducted by the BLS for major regions in
California. We specifically looked at survey data for the Los Angeles, Sacramento, and San
Francisco combined statistical areas (CSAs), which together account for about 75 percent of
Californias population. The surveys for Los Angeles and San Francisco have a reference month
of April 2010 (the mid-point of a 14 month data collection period), and the Sacramento survey
has a reference month of June 2010. The NCS is conducted by the BLS on an ongoing basis and
includes information on average hourly earnings for over 800 occupations and sub-occupations.
For our analysis, we focused on the subset of occupations that are displayed in the NCS surveys
for both the state and local government sector and private sector. Where possible, we further
narrowed the comparisons to standardized job levels (for example, a Level 3 administrative
assistant or a Level 9 manager) within occupations to minimize variations in wages due to
differences in job duties and complexity between the two sectors.
5
!Appendix tables 1 through 3
provide the detailed comparisons. Our main conclusions are:
As shown in Figure 3 below, state and local government sector pay is higher than private
sector pay in 21 out of the 30 occupational categories and subcategories we compared in
the Los Angeles CSA. Within the 19 occupations for which there were specific job levels
identified, state and local government sector pay was higher in 13 of the cases. State and
local government sector pay was also higher in 14 of 18 categories in San Francisco, and
8 out of 14 categories in Sacramento. !

4
Source: Current Population Survey, data for California households, 2009. U.S. Census Bureau.
5
Job levels are based on system that looks at four job-related factors knowledge, job controls and
complexity, contacts, and physical environment. Points for the four factors are recorded and totaled. BLS
publishes data for 15 job levels. For a more detailed description, see Guide for Evaluating Your Firms Jobs
and Pay. BLS . http://wwwbls.gov/ncs/ocs/sp/ncbr0004.pdf.!


Figure 3
Number of Occupations with Higher Hourly Wages
National Compensation Surveys, 2010

State and local government sector wage premiums are most pronounced in lower-skilled
occupations. For example, the state and local government sector premium in Los Angeles
for level-3 food preparation and serving jobs is over 28 percent, and the margin in level-3
building and grounds occupation is 26 percent.
Private sector pay premiums are mainly found in top-level management and specialized
occupations, such as engineering and computer science.
More Pay Variation in Private Sector
Expressed another way, pay rates are more compact in the public (state and local government)
sector. Figure 4 provides an example for engineers in the Los Angeles CSA. It shows that while
average hourly wages are similar in the public and private sector, the bottom 10 percent of public
sector employees are paid $9 per hour more than their private sector counterparts, while the top-
10 percent of public sector employees are paid $20 per hour less than their counterparts.


!
Figure 4
Comparison of Pay Dispersion for
Engineers in Los Angeles CSA
(Average Hourly Wage)

The greater private sector wage dispersion has important implications for public-private sector
wage comparisons. It suggests that differences in averages may have only limited applicability to
many of the workers within an occupational group. For occupational groups with similar
averages, less skilled workers in the state and local government sector are likely to receive higher
wages than their counterparts, while top employees are likely to be paid less.
California Department of Personnel Administration Survey
One limitation of the NCS is that it combines all state and local employees into one group, and
makes no distinction between the two levels of government. Salary surveys taken by the
Department of Personnel Administration (DPA) during the past five years suggest that there are
indeed significant differences in salary levels between the two levels of government.
Specifically, in a wage and benefits survey release in 2006, DPA compared salaries earned by state
employees to local governments for 41 benchmark job classes. It also made comparisons to
private sector salaries for 20 of 41 of these benchmark classes. The results of this survey are
included in Appendix Table 4. It showed that state government pay lagged the private sector in
12 out of 20 occupations, including all medical, executive, and managerial classes. It led private
sector pay in 8 occupations, mostly in trades and lower skilled occupations. It also found that local
government salaries led the state in 15 out of 20 occupations, and led the private sector in 14 out
of 20 occupations.
DPA also notes that actions taken since 2006 have narrowed the state pay gap in a few areas,
particularly in health related occupations within the Department of Corrections (due to federal
court-ordered increases). However, given the lack of general pay increases in recent years, it
would appear unlikely that the state has closed the gap in most occupations.


Limitations of the NCS and DPA Wage Surveys
While the wage surveys cited above provide direct pay comparisons for certain occupations, they
are subject to several qualifications.
First, they only apply to the subset of jobs that are present in both sectors. This is a
significant limitation, since well over one-fourth of all occupations are unique to one
sector or another. Police and firefighters are two examples of state and local government
sector jobs that do not have direct private sector counterparts. Indeed, some of the more
controversial elements of public sector pay involve compensation for safety employees (see
nearby box). Retail sales occupations are examples of jobs unique to the private sector.
Even in cases where there was overlap, our comparisons were limited to just those sub-
occupational categories for which earnings estimates for both sectors were available.
Finally, the results are not weighted for the number of workers in each comparison group.
For these reasons, the comparisons are best described as indicators of relative wage trends
as opposed to statistically valid measures of differences.
Second, the wages in these surveys are for straight time and do not include supplemental
pay for bonuses, profit sharing, or stock appreciation rights. The exclusion of these items
biases downward the pay for the private sector, where these forms of compensation are
provided. The overall amount of this bias is relatively modest, accounting for around 2
percent of pay for the private sector overall. However, the exclusion has more
pronounced effect on higher-level occupations, particularly for top management and
financial occupations.
The exclusion of premium pay for overtime affects the hourly earnings shown for both
public and private sector employees. It is a major factor in public safety classifications.
However, these classifications are not among the occupations shown in Appendix Tables 1
through 3 (due to lack of comparability between public and private sector jobs in this area).
The relative impact of overtime premium pay in the categories shown is probably minor.
One other issue worth noting is that is that the comparisons we are citing are for all full-time
employees, including those working for both small and large companies. If the private sector
comparison group were just employees of large companies, relative pay rates would look
somewhat less favorable for state and local government workers. The NCS data by firm size is
limited to broader occupation categories, so it is not possible determine exactly how much less
favorable, but we estimate it could be in the general range of 10 percent for comparable jobs.
We do not believe the comparison to all full time private sector employees creates a bias in the
results. We are noting this issue, however, because public-private sector wage comparison studies
often limit the private sector comparison group to employees of large sized firms, on the grounds
that state and local government sector workers are employed by large organizations. We discuss
this issue more fully in the following section.
Job Security Greater In Public Sector
A major issue in wage comparisons is the notable difference in job security in the public versus
private sector. Although job layoffs are currently more prominent in the state and local sector
than in the past, the overall risk to a civil service employee of an involuntary job separation is
substantially lower than a worker faces in the private sector. As shown in Figure 5, during the
2000 through 2010 period the rate of involuntary job separations (layoffs and firings) averaged
about 6 percent in the state and local sector and 20 percent in the private sector.
6

6
Source: Job Openings and Labor Turnover Survey. BLS. http://www.bls.gov/jlt/.


!
Using the historical difference between public and private sector involuntary separation rates, and
making conservative assumptions about duration of unemployment following an involuntary
separation, we estimate the greater security translates into a risk-adjusted wage premium for
public sector employees of over 3 percent.
7

Wages for Public Sector Safety Employees
One of the key limitations relating to wage comparisons between public and private sector jobs is
the number of occupations that are unique to one sector or another. Key examples are public
safety jobs such as peace officers, firefighters, and correctional officers -- for which there are no
direct private sector counterparts.
8

Though it is not possible to directly compare pay of public safety occupations to the private
sector, it is possible to make comparisons to other public agencies. In these comparisons,
California pay rates are well above the national average. According to BLS national
compensation survey data, hourly pay for police officers working in Los Angeles was one-third
higher than the national average in 2009 ($37 versus $27 per hour). Pay for correctional officers
(state prison and county jail guards) was more than 50% above the national average, and pay for
firefighters was 22% higher. Los Angeles is a relatively high cost, high wage area, but even after
accounting for this factor, the pay margin over the national average for state and local
governments is significant.
The BLS data does not provide direct comparisons for federal versus state/local sector safety
occupations. Part of the challenge is that the federal and state occupations have different
background requirements and job duties. In general, pay ranges are broader at the federal level,
so federal workers would appear to have more upward potential in at least some occupations. As
one example, the pay range for FBI agents in Los Angeles is between $72,000 and $148,000,
while a comparable range for an LAPD detective is from $80,000 to $111,000. However, the FBI
agents pay already includes an adjustment for assumed overtime (law enforcement availability
pay). In contrast, an LAPD detective has historically received additional overtime pay for hours
worked (though because of budget shortfalls, overtime pay has recently been limited).
This leads to a broader point related to overtime policies. They are more expansive at the state
and local level than at the federal level. In California, in particular, overtime pay has had
dramatic effects on public safety pay levels in some jurisdictions. For example, according to data
reported to the State Controllers Office, average pay reported on W-2s for fire captains in the
San Ramon Fire District was $173,000 in 2009 almost 70 percent more than the top end of the
published pay range ($103,000), with most of the difference due to overtime.

7
Assumes a worker is unemployed for four months following an involuntary separation, and that
unemployment insurance replaces of one-third of wages. As a point of reference, according to BLS data,
the median duration of unemployment as of March 2011 was just under five months, and the average was
over seven months. Estimate does not take into account losses of health insurance and other non-wage
compensation, which can be significant.
8
There are private sector jobs, such as private investigative services and protective services, which have
some elements in common with public safety jobs. There are also private sector jobs involving similar or
higher levels of risk (logging, transportation, or roofing). These private sector occupations generally pay
much less than public safety jobs. However, it is important to note there are substantial differences relating
to the obligations and responsibilities that sworn officers and fire fighters have in terms of protecting the
broader public.


Figure 5
Involuntary Separations Each Year: State and
Local Government Versus Private Sector
(Percentage of Workforce)

Statistical Approaches to Comparing Pay
A way in which researchers have attempted to address the lack of comparability of occupations
between the public and private sector is to take an alternative approach that focuses on people
rather than occupations.!This statistical approach has formed the basis for several recent studies
by state and local government sector advocates asserting that state and local workers are
undercompensated compared to the private sector.
The approach is a regression-based statistical analysis that uses either census data or current
population survey (CPS) data collected by the U.S. Census Bureau for the BLS.
9
The studies
make statistical comparisons of wages reported by individuals working in the two sectors, by
controlling for major earnings determinants, such as educational attainment, experience, broad
occupation, hours worked, and a variety of demographic characteristics (including sex, marital
status, race, and citizen status). After standardizing for all these earnings determinants, the
remaining difference in observed earnings between public and private sector employees is
assumed to represent the state and local government sector premium or shortfall.
Figure 6 shows the results of recent studies for California. All three of the studies shown
including the study completed by the conservative-leaning Heritage Foundation estimate
significant wage shortfalls for state and local government sector employees, ranging from
8.9 percent to 10.2 percent for state employees and 0.6 percent to 6.1 percent for local employees.
We note that while the Heritage Foundation comes to similar conclusions as the other studies

9
The CPS is a monthly survey of about 50,000 U.S. households. The objective of the survey is to create a
representative sample of U.S. households that provides a detailed picture of the demographic and economic
characteristics of people living in the U.S. and its various subdivisions.!


!
with respect to comparative wages, it reaches markedly different conclusions about the relative
levels of total compensation between the public and private sectors.
10

Figure 6
Recent Statistical Studies Showing Wage Shortfalls in California
State and Local Government Relative to the Private Sector\a
Study
Estimated
State
Government
Shortfall
Estimated Local
Government
Shortfall
Estimated
Combined
State/Local
Shortfall
National Institute on
Retirement Security,
April 2010
-9.8 % -6.1% NA
Center on Wage and
Employment Dynamics,
October 2010
-8.9% -5.4% -6.4%
Heritage Foundation,
March 2011
-10.2% -0.6% -3.7%
a\ See Appendix Table A-5 for citations and descriptions of these studies.
Criticisms of Statistically Based Wage Studies
The statistical approach to making standardized wage comparisons is subject to two main
criticisms. The first is that it uses inputs (such as educational levels and experience) rather than
outputs for determining whether there is a wage gap between the public and private sector
employees. The premise of this methodology is that two individuals one in the state and local
government sector and one in the private sector, with the same general educational and related
attributes -- ought to be making the same amount of money, without regard for whether the two
individuals are, in reality, working in equivalent jobs. The approach does not take into account
substantial differences that can exist between public sector and private sector occupations in terms
of job security, responsibilities, expectations, productivity, and other factors.
The second criticism is that the analyses inappropriately limit the direct private sector comparison
group to employees of large companies, thereby biasing the comparisons in favor of the private
sector. As noted earlier, this limitation is significant since large private sector companies pay
higher salaries than their smaller counterparts. For example, according to the CPS survey for
2009, a California worker that has a Bachelors degree and is employed by a firm with more than
1,000 employees, earns, on average, 13 percent more than his or her counterpart working in a
firm with less than 100 employees.
11
The results from the CPS survey are consistent with the
occupational survey data discussed above, which also show that larger firms pay more.

10
Specifically, the studies prepared by the National Institute on Retirement and the Center for Wage
Employment Dynamics conclude that state and local employees are not overcompensated when both wage
and non-wage benefits are accounted for. However, the study by the Heritage Foundation concludes that
total compensation is significantly higher in the state and local government sector. It reaches this conclusion
by including employer costs for accruals of retiree health benefits and by conferring a value for job security
in the state and local government sector.
11
Source: Current Population Survey, data for California households, 2009.


The rationale for limiting the comparisons to private sector employees of large firms is that,
because most state and local employees work for large employers (the government) they have
characteristics and job preferences similar to workers in large sized companies. The counter
argument is that, by limiting the comparison group to just employees of large firms, the state and
local government sector employees are being compared to a select group, since large firms tend to
be industry winners that can afford to pay more to attract top employees. This is of particular
importance in California, given that it is home to Google, Apple, and several other high-tech
companies that are among the most successful in the world.
To provide an indication of what the above results would look like if the comparison group were
all private sector employees, instead of just those working for large firms, we developed our own
regression-based estimates of wage differences using detailed CPS data for California.
12
We
followed the general approach used in the CPS-based studies cited above, using data drawn from
the 2006 through 2010 Annual Demographic March Supplement of the CPS. Under the baseline
scenario, we included firm size as an explanatory variable. Under the alternative, we excluded the
firm size variable from our regression equation.
As shown in Figure 7, our estimates show that state and local employees earn about 3.8 percent
less than employees with similar attributes working for large private sector firms. When the
comparison group is changed to include all private sector employees, the 3.8 percent penalty for
state and local government combined turns into a 3 percent wage premium.
Figure 7
Calculations of State and Local Government Sector
Wage Premium/Shortfall Using CPS Data For California
State and Local
Government Sector
Group
State and Local Government Sector Wage
Premium(+)/Shortfall(-) Compared To:
Employees of Large
Companies
All Private Employees
State Workers -8.6% -0.1%
Local Workers -1.3% +4.9%
Combined -3.8% +3.0%

Other Issues
The CPS survey data used for these comparisons has some advantages over the occupational
survey data discussed above. For example, the wage totals include overtime and other forms of
supplemental pay. However, the CPS data has its own limitations. First, it is based on self-
reported income, thus it is probably not as accurate as occupational surveys.
Second, the estimates may understate wages per hours worked for full time employees that work
for less than a full year. In particular, the CPS survey counts teachers as full time employees even
though their contracts call for a work year that is 38 weeks or less after taking into account the
summer break and school holidays. Because of this, the implied hourly wage (for hours actually
worked) is understated by nearly one third. This distortion is significant because teachers account

12
We developed the regression estimates using data from Integrated Public Use Microdata Series, (IPUMS)
Current Population Survey: Version 3.0. Minneapolis: University of Minnesota, 2010.


!
for nearly 14 percent of the total state and local government employment base. We believe that
properly accounting for time worked would reduce the state and local government wage gaps
cited above by between 1 percent and 2 percent.
Finally, the CPS survey is too small to allow for meaningful comparisons of jobs within sub-
regions of the state. This is significant, given the variation in cost of living and wage rates that
exists in different regions of California.
Bottom line on Wage Comparisons
Although both are imperfect measures, the occupational surveys and statistical models point in
the same direction. They indicate:
When compared to all full time private sector workers, wages of state and local workers
are similar to, or slightly higher than, than wages for comparable workers in the private
sector.
When compared to just workers of larger private sector firms, state and local government
jobs pay levels appear to be a little less than the private sector.
State and local government sector pay is considerably higher in many less-skilled
occupations, and is lower in some high-skilled and specialized occupations (top level
management and computer specialists).
There is a significant pay gap between state government and local government, with local
government paying more in many occupational categories.
Within occupations there is much more wage variation in the private sector (or
alternatively, there is more wage compression in the state and local government sector).
Thus, for occupations with similar average wages, those at the top end of the occupations
pay range are likely to be paid higher in the private sector, while those at the bottom of
the range are likely to be paid more in the state and local government sector.
Non-Wage Benefits
While the exact relationship between state and local government sector and private sector pay is
open to some debate, there should be little question about non-wage compensation. The majority
of state and local employees, who receive both retiree benefits and health care, enjoy non-wage
benefit levels that exceed their private sector counterparts by a substantial margin. This is
particularly true for long-term employees with fully vested retirement benefits.
Comparisons between state and local government sector and private sector benefits involve two
issues. One is the incidence of benefits (that is, how likely is a given employee to receive the
benefit) and the other is the richness of the benefits for those who receive them. State and local
government employees are more likely to receive a full range of employee benefits than their
private sector counterparts, and the benefits they receive are likely to be richer than benefits
received by private sector employees.
Incidence of Benefits
Regarding the incidence of benefits, state and local government employees have moderately
greater access to employer-provided health care and paid leave, and substantially greater access to
retirement benefits. For example, according to the most recent NCS benefits survey, medical
health benefits are provided to 90 percent of state and local government employees, compared to
71 percent of private sector employees, and 89 percent of employees working for firms with more
than 100 employees. Dental care is provided to 84 percent of state and local government


employees, versus 54 percent of private sector employees and 71 percent of private sector
employees working for larger firms. Regarding retirement benefits, as noted in Chapter 1, over
80 percent of state and local government employees have access to defined benefit plans versus
about 20 percent in the private sector.
Richness of Benefits
Even when compared to the subset of large private sector companies offering a full range of
benefits, state and local government employees come out ahead. Our review indicates that the
state and local sector offer modestly more paid leave days, and health plans that appear to be
richer, on average, with respect to plan designs and cost-sharing. The major differences, though,
are in retirement programs. The public sector margin is substantial, particularly for long-term
employees.
The differences are also much greater than recent estimates would indicate. Most recent studies
rely on BLS data on employer costs for employee compensation, which we believe understates the
margin for state and local government employees for three main reasons:
First, the BLS data on compensation costs excludes most costs related to retiree health
care. The expenses are not recognized in the BLS survey unless employers are prefunding
the benefits during employees working years, which rarely occur in the public sector.
The accrual of retiree health care is clearly a form of compensation, regardless of whether
it is paid for up front or in the future.
Second, as noted in Chapter 1, California public pension funds are facing large actuarial
shortfalls. Recent contributions have been much less than the amount needed to cover
the true costs of employee pension benefit accruals.
13

Third, more stringent accounting and funding rules apply to private sector pension funds.
These rules require private sector employers to make comparatively larger annual
contributions to finance a given level of future benefits. The different funding rules have
no impact on the benefit accruing to the employee. They merely affect the amount of
these benefits that must be paid for today versus in the future.
An Apples-To-Apples Comparison of Public
and Private Sector Benefits
To provide a more direct comparison of benefits being committed to government versus private
sector employees, we calculated benefits for a typical mid-career state worker. Our example is a
45-year old employee that is one-half way through a 30-year career. We compare the wages and
benefits earned by that employee to those provided to an individual with the same characteristics
but employed by a typical large private sector firm.
As a starting point, we assume the state worker makes $60,000 -- or $5,000 less than his or her
private sector counterpart. For the estimates of state benefits, we relied on recent bargaining
contracts for miscellaneous state employees. The private sector estimates are based on the average
retirement benefits offered by the sample of large private sector firms used in our Chapter 1
retirement comparisons, as well as information from surveys of private sector companies by the

13
As one example, the actuarial valuation released in March 2011 for CalSTRS, the states second largest
pension fund, found that it faces an actuarial shortfall of $56 billion. Elimination of the shortfall over a 30-
year period would require an immediate increase in employer contributions of $3.9 billion annually
which would require additional funding equal to 14 percent of teachers salaries.


!
BLS National Compensation Survey, the U.S. Chamber of Commerce, and the Kaiser Family
Foundation.
14

Our estimates of annual retirement benefit accruals are based on a present value calculation of
the employer-provided portion of pension and retiree health benefits earned by the employees
over their full 30-year careers. The amount for the state worker assumes the CalPERS 2% at age
55 retirement formula that is in effect for miscellaneous state employees hired prior to 2011.
For purposes of this calculation, we assume the benefits are earned proportionally over the
workers career.
Results
As shown in Figure 8, non-wage benefits are considerably higher for the state worker than the
private sector counterpart. Specifically, the state worker would receive $46,492 in non-wage
benefits, resulting in total compensation of $106,492. The private sector worker would receive, on
average, $31,737 in non-wage benefits, resulting in total compensation of $96,737.
The state employee receives modestly more in health care and paid leave, and slightly less in
Social Security and Medicare (because of lower wages). The main source of the overall difference,
however, is the value of pension and retiree health care. The combined retirement benefit would
be slightly over $19,100 per year for the state worker, compared to the slightly over $5,700 for the
private sector worker. For the state worker, the accrual of retiree health care benefits accounts for
about $8,000 and the pension benefit is worth about $11,000.
Figure 8
Comparison of Benefits: State Employee Versus
Typical Employee of Typical Large Private Sector Firm
Category of Compensation State Employee
Private Sector
Employee
Wages $60,000 $65,000
Benefits - -
Employer-paid health $12,381 $11,475
Paid leave $10,385 $9,570
Social Security & Medicare $4,590 $4,972
Retiree pension & health $19,136 $5,720
Total Compensation $106,492 $96,737

a\ Benefits for both state and local government sector and private sector active and retiree health plans
are a weighted average for employees with zero, one, and two-or-more dependents.

14
Sources: National Compensation Survey, Bureau of Labor Statistics. Employer Health Benefits 2010
Annual Survey, Kaiser Family Foundation and Health Research and Educational Trust. 2008 Employee
Benefits Study, U.S. Chamber of Commerce. The private sector comparison group for retirement benefits
includes Chevron Cisco, McKesson, Northrop Grumman, Qualcomm, and Safeway.


Caveats
A significant portion of the difference in the example is due to the high value associated with the
retiree health care provided to State of California employees. The benefit comparisons would less
favorable for a public employee working for a local agency that had considerably less generous
retiree health benefits (though it is important to recall that local governments pay higher wages
and some have higher pension benefits than the state). It would also be less favorable for members
of CalSTRS, who, as shown in Chapter 1, receive less rich retirement benefits than other public
sector employees.
It is also the case that the value of retirement benefits in a pension system is considerably greater
for an employee working a full career than for an employee that terminates service after 10 or
15 years. This is particularly true for employees that terminate prior to fully vesting in retiree
health care.
However, the context for pension reform discussions is the unsustainable benefits being offered to
long-term employees and the need to reduce them in order to rein in government costs. To the
extent it is mainly long-term employees whose pensions would be affected by reforms, it is
appropriate to look at compensation comparisons for such full career employees.
Conclusions
Our analysis of occupational survey data and statistical studies finds that average wages in the
state and local government sector combined are roughly similar to private sector average pay
levels for comparable jobs a little above the average of all private sector workers and a little
below private sector workers employed by large firms. The surveys and statistical analyses indicate
that state government pay lags behind that of local government. The surveys also reveal that there
is much greater wage variation within the private sector, so that average wages within occupations
do not tell the full story.
It is also important to recognize the limitations of these wage surveys and studies. They do not,
for example, attach any value to the relatively greater job security in the public sector, nor do they
pick up other factors that might attract an individual to one sector or the other.
Employee benefits are more prevalent in the state and local government sector than the private
sector, and the retirement benefits are considerably richer for long-term public sector employees.
These higher benefits raise compensation for long-term state and local government employees by
a substantial margin, putting them ahead, on average, of their private sector counterparts. This is
particularly true for state and local employees covered by retiree health care.
Finally, from an economic and policy perspective, the key question is not who has the higher
compensation levels but rather what is an appropriate pay and benefit package for attracting and
retaining a qualified workforce in each sector. An equally important question is whether the
compensation systems have enough flexibility built in to respond to rapidly changing economic
and budgetary circumstances. A key rationale for pension reform is that this flexibility is lacking in
the state and local government sector. The current public compensation systems are
overcommitted to large vested pension rights, which do not provide state and local governments
with adequate flexibility to manage their budgets.

Chapter 2: Appendix Tables
Appendix Tables 2-1 through 2-5 provide various detailed information on employee public and
private sector employee compensation.
Tables 2-1 through 2-3:
BLS Occupational Wage Comparisons
Tables 1 through 3 present occupational wage comparisons based on data from the BLS National
Compensation Survey. The hourly wage comparisons are for all full-time workers, are for straight
time, and do not include overtime premiums or supplemental pay.
Table 2-1 provides the comparisons for the Los Angeles consolidated statistical area
(CSA), which covers the counties of Orange, Riverside, San Bernardino, Ventura, and
Los Angeles. These counties have a population of about 18 million, representing slightly
less than one-half of the statewide total.
Table 2-2 provides comparisons for the San Francisco CSA , which includes the counties
of San Francisco, San Mateo, Santa Clara, Alameda, Contra Costa, Napa, Marin,
Sonoma, and Solano. These counties have a combined population of 7.4 million,
comprising about one-fifth of the statewide total.
Table 2-3 provides comparisons for the Sacramento CSA, which includes the California
counties of Sacramento, Yolo, El Dorado, Placer, Sutter, Yuba, and Nevada. (This CSA
also includes Douglas NV, but the overwhelming majority of jobs covered in this CSA are
attributable to California.) The California counties have a combined population of
2.4 million, which comprise about 6% of the statewide total.
Comparisons for Specific Job Levels
Where possible, we compare public versus private sector occupational wages for specific job
levels. These levels have been designed by BLS to create a consistent standard for wage
comparisons and for comparison across occupations.
In order to establish these levels, the BLS has created a system involving scoring for four
occupational factors knowledge, job controls and complexity, contacts (nature and purpose),
and physical environment. A job is assigned points for each of the four factors based on detailed
criteria set forth by BLS.
1
The points for each of the factors are added together to arrive at a
grand total, which is then converted into a job ranking of from 1 to 15. At the lower end of the
scale, (levels 1 through 4) are entry-level jobs requiring only a basic understanding of the
discipline. In such jobs, the worker is subject to significant job controls, follows pre-existing
procedures, may perform repetitive tasks, and has contacts mainly within the workplace that are
for the purpose of receiving directions. At the upper end of the scale (levels 11 through 15) are
employees that set policy based on little or no specific guidance from others, have an integral
understanding of the discipline (and in fact may contribute to the body of knowledge of that
discipline), and whose interactions are with high level business leaders or public officials, for the
purpose of influencing actions or policies. Mid-level employees generally fall in the range of from
level 6 to level 10.

1
For a detailed description, see Guide for Evaluating Your Firms Jobs and Pay. Bureau of Labor Statistics.
http://wwwbls.gov/ncs/ocs/sp/ncbr0004.pdf.


Highlights
The tables show:
Of the 30 occupations shown for the Los Angeles CSA, 21 have higher average wages in
the state and local government sector, and 9 have higher average wages in the private
sector. For the 19 occupations where specific job levels are identified, 13 are higher in the
state and local government sector and 6 are higher in the private sector.
Of the 18 occupations shown for the San Francisco CSA, 14 have higher average wages
in the state and local government sector, and 4 have higher average wages in the private
Of the 14 occupations shown for the Sacramento CSA, 8 have higher average wages in
the state and local government sector, and 6 have higher average wages in the private
Table 2-4: DPA 2006 Survey
Appendix Table 2-4 provides information from the California Department of Personnel
Administration Survey of state, local, and private sector job classifications in 2006. In the survey,
DPA developed salary comparisons between the state and local governments for 41 benchmark
classifications. It also developed comparisons for 20 of the 41 classifications for which there was a
private sector counterpart. Its main findings were:
The state of California lagged local governments in 15 out of 20 benchmark jobs.
The state of California led the private sector in 8 categories, but lagged in 12 benchmark
jobs.
DPA notes that the survey was taken prior federal court ordered pay increases in health care
occupations related to the Department of Corrections.
Table 2-5: Literature Review
Appendix Table 2-5 summarizes the results of our review of academic literature, government
surveys, and other information related to compensation comparisons between the public sector
and private sector. For each item, we include information on the purpose of the survey or study,
the approach it uses, and its major conclusions, along with key comments and criticisms.
The entries are grouped into six major categories: (1) aggregate compensation comparisons;
(2) more detailed occupational wage surveys; (3) statistical-based comparisons of wages in the state
and local government versus private sector; (4) studies on compensation growth over time;
(5) nuances of private-public comparisons; and (6) other resources relating to private sector
employee benefits.
For each entry we describe the study or resource, identify its purpose and the approach it uses,
highlight its main conclusions, and provide comments and criticisms.
Our review found relatively few peer-reviewed academic studies focused on California. Most
recent studies for California and other states have been prepared by organizations on one side of
the public-private sector compensation debate or the other. We have attempted to include
criticisms from both sides of the debate.

Appendix Table 2-1
Comparison of Average Hourly Earnings By Occupations
National Compensation Survey, Los Angeles CSA, April 2010

Occupation
State and
Local Sector
Private Sector Difference
Management:
Level 9 $36.94 $33.48 10.3%
Level 11 $48.89 $50.71 -3.6%

Business and Financial:
Level 7 $32.03 $25.69 24.7%
Level 8 $27.13 $25.54 6.2%
Level 9 $36.23 $35.79 1.2%

Management Analyst:
All $37.28 $40.64 -8.3%
Accountants and Auditors $36.71 $28.00 31.1%

Computer and Mathematical
Science:
All $35.69 $34.24 4.2%
Computer Support Specialists $28.50 $26.48 7.6%
Computer Systems Analyst $38.89 $38.18 1.9%

Architecture and Engineering:
All $41.04 $41.67 -1.5%
Engineer $46.92 $51.92 -9.7%

Legal $43.74 $37.26 17.9%

Education, Training, and Library:
Level 9 $49.13 $34.76 41.3%
Level 11 $57.28 $37.79 51.6%

Postsecondary Teachers:
Level 11 $57.83 $40.29 43.6%


Occupation
State and
Local Sector
Healthcare Practitioner & Technical:
Level 9 $38.48 $41.48 -7.2%
Level 11 $52.24 $49.62 5.2%

Registered Nurse:
Level 9 $38.24 $40.71 -6.1%

Healthcare Support:
All $14.66 $14.50 1.1%
Nursing, Home Care $13.44 $11.40 17.9%

Food preparation:
Level 3 $14.22 $11.04 28.9%

Building and Grounds:
Level 3 $16.66 $13.13 26.9%

Office and Administrative Support:
Level 3 $17.53 $13.75 27.5%
Level 4 $18.33 $16.77 9.3%
Level 5 $19.98 $20.06 -0.4%
Level 6 $22.52 $23.88 -5.6%

Installation, Maintenance and
Repair:
Level 6 $28.11 $27.12 3.7%
Level 7 $33.81 $36.10 -6.3%

Transportation and Moving $25.97 $15.25 70.3%

Appendix Table 2-2
Comparison of Hourly Earnings by Occupation
National Compensation Survey, San Francisco CSA, April 2010

Occupation
State and
Local Sector
Management $55.25 $51.93 6.4%

Business and Financial:
Level 9 $35.08 $34.67 1.2%
Level 10 $43.08 $34.64 24.4%

Computer & Mathematical Science:
Level 9 $35.89 $39.83 -9.9%

Engineer:
Level 9 $47.63 $39.60 20.3%

Life, Physical, and Social Sciences $34.99 $41.49 -15.7%

Community & Soc. Svc.
Occupations:

Level 7 $28.11 $18.20 54.5%

Legal $41.62 $62.24 -33.1%

Education, Teaching, Training:
Level 10 $49.64 $46.55 6.6%
Level 12 $79.41 $77.98 1.8%

Level 9 $52.56 $52.38 0.3%

Registered Nurse:
Level 9 $56.58 $53.85 5.1%

Health Care Support $21.55 $20.28 6.3%


Occupation
State and
Local Sector
Building and Cleaning:
Level 3 $17.12 $13.43 27.5%

Level 3 $17.06 $15.60 9.4%
Level 6 $25.33 $25.13 0.8%

Installation and Repair:
Level 5 $23.80 $29.81 -20.2%
Level 6 $31.85 $28.94 10.1%

Appendix Table 2-3
Comparison of Hourly Earnings by Occupation
National Compensation Survey, Sacramento CSA, June 2010

Occupation
State and
Local Sector
Management
Level 9 $32.68 $35.30 -7.4%

Business and Financial
Level 9 $30.43 $34.30 -11.3%

Computer Systems Analyst $37.76 $38.85 -2.8%

Education, Training, and Library
Level 9 $50.14 $33.03 51.8%

Level 9 $48.33 $47.66 1.4%

Registered Nurse $45.77 $49.69 -7.9%

Food Preparation $18.04 $10.95 64.7%

Janitor $15.93 $12.32 29.3%

Level 3 $16.37 $14.72 11.2%
Level 4 $17.19 $16.59 3.6%
Level 5 $18.66 $19.44 -4.0%
Level 6 $20.06 $21.65 -7.3%

Installation and Repair $26.94 $22.57 19.4%

Transportation and Material Moving $23.68 $19.76 19.8%


Appendix Table 2-4
Department of Personnel Administration Survey of Occupations
Average Monthly Wage

Occupation
State
Government
Local
Government
Private
Sector
Occupational Therapist $3,960 $5,900 $5,515
Pharmacist $5,748 $7,766 $7,970
Social Worker Masters Level $4,139 $4,611 $5,116
Respiratory Care Practitioner $3,616 $6,503 $4,454
Chief Financial Officer $10,951 $11,126 $13,290
Licensed Vocational Nurse $2,967 $3,755 $3,296
Director, Human Resources $10,271 $9,920 $11,384
Auditor $5,247 $5,129 $5,692
Attorney $7,386 $8,955 $7,845
Chief Information Officer $10,271 $11,126 $10,908
Programmer Analyst $5,247 $5,715 $5,550
Registered Nurse $5,423 $5,004 $5,691
Office Assistant $2,641 $2,879 $2,555
Electrician $3,926 $5,507 $3,778
Budget Analyst $4,997 $6,288 $4,763
Accountant $4,997 $5,244 $4,707
Personnel Analyst $4,997 $5,800 $4,507
Stationary Engineer $4,601 $4,474 $3,839
Custodian $2,382 $2,507 $1,851
Cook $3,021 $2,710 $2,292

95 | Appendix Table 2-5
Appendix Table 2-5
Summary of Literature Review Regarding Compensation
Aggregate Compensation Comparisons

Name/Source Purpose/Approach Conclusions/Comments
Employer Costs for
Employee
Compensation
December 2010
Bureau of Labor Statistics.
March 9, 2011
http://bls.gov/news.release/pdf/
ecec.pdf
Purpose:
To determine the average cost to
employers for wages and benefits per
employee hour worked.
Approach:
Data is drawn from BLS National
Compensation Survey. It is based on
sample of 62,400 occupations from
13,100 private employers and 11,600
occupations from 1,800 state & local
government employers.

Conclusions:
Average employer cost of compensation for private sector
employees is lower than for state/local employees
nationally it was $27.75 per hour for private sector
employees versus $40.28 per hour for state/local
government employees in 2010.
State/local employees receive a greater proportion of
compensation from benefits. Private sector employees
receive 29.2% of total compensation from benefits, versus
34.4% for state/local employees.
Comments:
Aggregate average compensation levels provide only
limited information about pay levels in the public versus
private sector, due to variation in the mix of occupations
between sectors.
The benefit measures are based on a survey of actual
expenditures for retirement benefits, rather than the
amount of benefits being committed to. To the extent
that public sector retirement plans underfunded, the
measures understate true employer costs of benefits in the
public sector.

Employee Compensation
in State and Local
Governments
Cato Institute. January 2010
http://www.cato.org/pubs/tbb/t
bb-59.pdf
Purpose:
To show that aggregate state and local
employee compensation is higher than
the private sector.
Approach:
BLS June 2009 Employer Costs for
Employee Compensation (ECEC) for
comparative costs of compensation. BLS
CPS 2009 March supplement for benefits
availability
Conclusions:
Average compensation is 45% higher for state/local
employees than private employees (no controls)
nationally; 59% higher for Pacific region (includes CA)
A substantially higher proportion of state/local
government employees have health insurance, retirement
benefits, life insurance, and paid sick leave than private
sector employees.
Public sector compensation is better and will grow even
more generous.
Comment:
Average compensation levels do not take into account
major differences in the compensation of public versus
private sector jobs, as noted above.

Occupational Survey Comparisons

Name/Source Purpose/ Approach Conclusions/Comments
National Compensation
Survey
Bureau of Labor Statistics.
Ongoing

Purpose:
To determine average hourly pay of
workers in numerous occupational
categories.
Approach:
Detailed occupational surveys
conducted by the BLS Office of
Compensation Levels and Trends.
Includes information for private sector
employees and state and local
government employees.
Occupational survey data available
nationally, regionally, and for
metropolitan statistical areas (including
7 in California).
Surveys include data on earnings by job
level within occupations, using a point
factor leveling system. The job factors
are knowledge, job controls and
complexity, contacts, and physical
environment.
Conclusions:
For the subset of occupations present in both sectors,
recent comparisons for MSAs within California show
state and local wage premiums in the majority
occupations that are present in both public sector and
private sector surveys.
State and local premiums most likely to be found in lower
paying and less skilled categories. Private sector
premiums mainly in high skilled and specialized
occupations (management, engineering, computer
specialists.
State and local government pay levels more compressed
than in private sector.
Comments:
Comparisons based on base pay, and do not include
overtime premiums and supplemental bonuses or profit
sharing payments.
Public-private sector comparisons hampered by large
number of occupations unique to each.
Name/Source Purpose/ Approach Conclusions/Comments
Total Compensation
Survey
CA Department of Personnel
Administration. April 2006
http://www.dpa.ca.gov/tcs2006/con
tents.htm
Purpose:
To compare CA state employee
compensation to employees of local
governments and the private sector.
Approach:
Looked at CA state employee
compensation data from 2006 survey of
benchmark classifications, including
34 journey-level classes and 7 executive-
and managerial-level positions.
Excluded are supervisory classifications
and peace officer/firefighter classes.
Private employee compensation data
aggregated from five professional
organizations 2005 surveys. Twenty
job classifications found comparable.
Did not collect data on private sector
bonuses and incentive pay programs.
Conclusions:
State employees were paid, on average, less than private
sector employees in 12 of 20 classifications, and more
than private sector employees in the remaining 8 (mostly
lower skilled) classifications.
Local government employees were compensated more
than private sector employees in 14 out of 20
classifications.
State compensation lagged local government employers
surveyed in 15 out of 20 classifications; in most job
classifications this lag is between 15 - 30%.
Comment:
State compensation levels in some health related classes
have increased substantially since the survey was
completed, due to court-ordered increases in pay.

Statistical Based Comparisons

I. Studies Asserting that
state and local government
employees are underpaid.
The Truth about Public Employees in
California: They are Neither
Overpaid nor Overcompensated.
Center on Wage and Employment
Dynamics (CWED), UC Berkeley.
October 2010
http://www.irle.berkeley.edu/cwed/
wp/2010-03.pdf
Debunking the Myth of the
Overcompensated
Public Employee. Economic Policy
Institute. September 2010.
http://epi.3cdn.net/8808ae41b08503
2c0b_8um6bh5ty.pdf.
Desperate Techniques
Used to Preserve the Myth of the
Overcompensated
Public Employee. Economic Policy
Institute. March 2011
http://epi.3cdn.net/1e05db309d0aa6
4571_rxm6bngw8.pdf
The Wage Penalty for State and
Local Government Employees.
Center for Economic and Policy
Research. May 2010
http://www.cepr.net/documents/pub
lications/wage-penalty-2010-05.pdf
Out of Balance? Comparing Public
and Private Sector
Compensation over 20 Years.
National Institute on Retirement
Security. April 2010
http://www.nirsonline.org/storage/ni
rs/documents/final_out_of_balance_r
eport_april_2010.pdf
Purpose:
To demonstrate that California
employees are not overcompensated
compared to their private sector
counterparts.
Approach:
To compare wages, these studies use
census and/or annual Current
Population Survey (CPS) data to
develop regression-based statistical
models, which relate employee wages to
a set of human capital (education and
experience), demographic, and
economic related attributes.
The models then attempt to measure
whether public sector and private sector
employees having similar attributes are
paid at different levels. Specific studies
vary in terms of data samples and
model specifications.
To compare benefits, the studies
generally rely on data from data in the
BLS National Compensation Survey or
its related series on employer costs for
employee compensation (ECEC).
Conclusion:
Depending on the study, the authors assert that total
compensation for state and local employees combined is
either roughly equal to or modestly less than the private
sector. They claim that higher public sector benefits are
more than offset by lower public sector wages.
Studies for California show state government employees
lagging local government employees with respect to
wages.
Criticisms/Comments:
Evaluations based on broad characteristics of
employees, not the jobs they do.
Results are sensitive to specific model specifications.
Most the studies limit private sector comparison group
to employees of large firms, which is controversial.
Treatment of teacher pay (12 months pay for 9 month
school year) may distort public sector results.
Comparisons undervalue public sector retirement
benefits for current employees, thereby seriously
understating public sector compensation.
See Heritage Foundation and Center for Union Facts
entries below for more detailed criticisms.
II Studies asserting state
and local employees are
over compensated:

Are California Public Employees
Overpaid?
Heritage Foundation. February
2011.
http://www.aei.org/docLib/Are-
California-Public-Employees-
Overpaid.pdf
The Economic Policy Institute Is
Wrong: Public Employees ARE
Overpaid . The Center for Union
Facts. February 2011.
http://www.unionfacts.com/downloa
ds/Public_Sector_UnionsBrief.pdf
Public Sector Unions and the
Rising Costs of Employee
Compensation. Cato Journal.
Winter 2010
https://www.socialsecurity.org/pubs/
journal/cj30n1/cj30n1-5.pdf

Purpose:
To demonstrate that state and local
workers are over compensated.
Approaches:
Heritage Foundation and the Center for
Union Facts studies followed same
human capital regression approach as
described in the previous panel, but
used different data sets and different
model specifications. The Center for
Union Facts broadened the private
sector comparison group so it included
workers of all-sized firms.
The Cato study used compensation and
employment data from BEA Regional
Economic Accounts for 2008 and data
on unionization rates from the CPS
survey for 2009.
It developed regression equations
relating the intensity of unionization to
public sector private sector wages by
state.

Conclusions:
Heritage Foundation study concluded that, public sector
wages lagged private sector pay by a modest amount,
total compensation in the public sector was much higher
due to non-wage benefits (retirement pensions and
retiree health) and greater job security.
The Center for Union Facts study found that when the
private sector comparison group is broadened to include
all private sector employees) state and local government
wages are modestly higher than the private sector in
California.
Cato study found that public sector unions increase the
compensation of the state/local government workforce
by 8.1% on average.
Criticisms/Comments:
Outputs are sensitive to model specification.
The Heritage Foundation Studys assertion about the
value of job security is controversial, since economic
studies on the relationship between industry wages and
job security are inconclusive.
Many factors besides unionization rates affect employee
pay.
See Desperate Techniques Used to Preserve Myth of
Overcompensated Employees. ECI, March 2011 for
these and other criticisms.
III Other econometric
based studies:
A National Analysis of
Public/Private Wage
Differentials at the State and Local
Levels by Race and Gender
Gregory B. Lewis
Andrew Young School of Policy
Studies at Georgia State University
http://aysps.gsu.edu/files/11-
10_LewisGalloway-AnalysisofPublic-
PrivateWageDifferentials.pdf
Purpose:
To determine whether state and local
employees are overcompensated or
undercompensated relative to private
sector employees. Also to determine
whether the relationship between public
sector and private sector compensation
has changed over time, and to
determine the effects of race and gender
on the comparisons.
Approach:
Used data from the 1990 and 2000 Census
and the 2005 and 2006 American
Community Surveys. Ran three regressions
using varied statistical techniques.
Conclusion:
Mixed results, with two of the three regression
techniques showing an earnings lag for state and local
employees, and the third showing a small surplus for
state and local employees.
The one regression run for California showed a 1%
wage premium for employees of state and local
governments during the 2001-through 2006 period.

Compensation Growth

Employer Cost Index.
Historical Listing. Bureau
of Labor Statistics. March
2011
http://www.bls.gov/web/eci/ecco
nstnaics.pdf

Purpose:
To determine how compensation in the
public (state and local government) and
private sector has compared over time.

Approach:
Used historical data from BLS National
Compensation Survey. Benefits covered
by the survey are:
Paid leave
Supplemental pay
Insurance benefits
Retirement and savings benefits
Legally required benefits
Other benefits (severance pay
and supplemental
unemployment plans)
Conclusions:
From 2001-2010, state/local government employees
compensation grew faster than private sector employees
(8.6% for state/local government employees versus
3.8% for private sector employees).
Most of the differential is in benefits.
Earnings growth is nearly identical - 1.8% for
state/local government employees versus 1.1% for
private sector employees. Benefit growth is quite
different - 24.9% for state/local government employees
versus 10.5% for private sector employees.
Comments:
The Center for American Action Fund indicates that
over the longer term (1991 through 2010), the public
sector and private sector growth rates are nearly the
same.
(http://www.americanprogressaction.org/issues/2011/
03/pdf/statebudgetissuebrief.pdf)

Nuances of Private-Public Wage Comparisons

Name/Source Approach Conclusions/Comments
Public-sector wage
comparability: the role of
earnings dispersion
Dale Bellman
Michigan State University
John S. Haywood
University of Wisconsin
Milwaukee
Public Finance Review 32(6): 567-
587. November 2004
https://pantherfile.uwm.edu/heywoo
d/www/567.pdf 567.full.pdf+html
Purpose:
To determine if average wage
differentials between the private
sector and public (state and local
government) sector is an appropriate
measure for comparability.
Approach:
Uses BLS CPS May 1993 sample of
7,897 private-sector workers, 409
federal workers, 458 state workers,
and 779 local workers.
Regression of pay differentials by
sector, with controls for education,
age, region of the country, marital
status, union status, race, urban
residency, broad occupation, job
tenure, part-time status, and
establishment and firm size.
Averages the absolute values of wage
differentials. This avoids the
cancelling out of overpaid and
underpaid workers.
Conclusions:
Public-sector earnings show less dispersion than private
sector earnings.
Individual earnings differentials favor the public sector at
the bottom of the earnings distribution and the private
sector at the top of the distribution.
Only 17.8% of state and 26.9% of local government
employees earnings are within a 5% range of private
sector counterparts, even though differences in averages
are within the range of 5%.
This suggests that average wage differences between
sectors provides only limited information regarding the
relative pay for many of the workers within the sectors.
The wage structure and the
sorting of workers into the
public sector
George J. Borjas
NBER. October 2002
http://www.nber.org/papers/w9313
Purpose:
To determine how wage structures
have changed from 1960 to 2000 in
the public and private sectors.
Approach:
Data sample drawn from the Public
Use Microdata Samples (PUMS) of
the U.S. Census 1960 - 1990 and the
March Annual Demographic
Supplement of the BLS CPS 1977
2001.
Ran regression relating weekly wages
to educational attainment, age, race,
and region of residence; estimated
separately by year, sector, and gender
group; pay gap determined by
difference between predictions for
public and for private in a given year
Conclusions:
Over 1960-2000, public sector male workers were paid 5-
10% less than private sector counterparts.
Public sector female workers earnings were at an
advantage in 1960 but declined to parity by 2000.
In 2000, state/local government male employees earnings
were 12%/10% less than private sector counterparts.
State/local female employees earnings 11%/5% less than
private sector counterparts.
Wage dispersion in public sector has decreased.
State and Local Pensions
are Different Than Private
Plans
Center for Retirement Research
Alicia Munnell and Mauricio Soto.
November 2007
http://crr.bc.edu/images/stories/Brie
fs/slp_1.pdf
Purpose:
To identify key differences between
public sector and private sector
pensions.
Approach:
Analyzed data from U.S. Board of
Governors of the Federal Reserve
System (2007), U.S. Census Bureau
(2007), and U.S. Department of
Labor (2007)
Conclusions:
Pension assets per worker are $185,900 for state/local
sector, $84,800 for the private sector. This disparity is
because coverage and accrual rates are both higher the in
public sector.
Public sector employees less likely to have social security.
Both employee and employer contributions to pensions
are higher rates for state/local than private sector.
Choosing Public Sector
Employment: The Impact
of Wages on the
Representation of Women
and Minorities in State
Bureaucracies
Jared J. Llorens, Jeffrey B. Wenger,
and J. Edward Kellough.
J Public Adm Res Theory 18(3): 397-
413. September 2007
http://jpart.oxfordjournals.org/conte
nt/18/3/397.full

Purpose:
To determine how wage differentials
affect the overrepresentation of
women and minorities in state
bureaucracies.
Approach:
Used date from CPS data for 1987,
1994, and 2002
Ran separate wage regressions for
each state in each year. Equations
related wages to age, education,
marital status, occupation, and
industry, as well as variables for sex,
ethnicity, and public or private sector
employment.
Conclusions:
Found that above average representation of women and
minorities in state government employment is associated
with small wage premiums paid in the public sector versus
private sector.

Other Resources

2008 Employee Benefits
Study
U.S. Chamber of Commerce .
April 2009
Purpose:
To identify benefits provided by
employers to employees in the U.S.
Approach:
Survey of 265 U.S. employers in
2007, representing a cross section of
all U.S. employers.
Conclusions:
More than 89% private companies provided vacation,
holiday retirement, and health insurance benefits to full
time employees.
The average cost-per-employee of these benefits was
$14,919.
The cost of benefits averaged about 29% of payroll, with
considerable variation among companies.
Larger companies offer a greater number of employee
benefits than do smaller companies.
Employer Health Benefits
2010 Annual Survey. March
2011
Kaiser Family Foundation and
Health Research and Educational
Trust

Purpose:
To identify health benefits provided
by U.S. employers.
Approach:
The 2010 survey included 3,143
randomly selected public and private
firms with three or more employees
(2,046 of which responded to the full
survey and 1,097 of which responded to
an additional question about offering
coverage).
Conclusions:
Detailed information about medical benefits provided by
employers in 2010. Some highlights:
68% of small firms and 99% of large firms offered health
benefits.
Total premiums averaged $5,049 for single coverage and
$13,770 for family coverage.
Employee share of premium costs rose in 2010, after
several years of stability.
The share of worker contribution for premiums averaged
$899 for single coverage and $3,997 for families.


Chapter 3: Government Cost of Retirement Programs 107
Chapter 3:
Government Costs of Retirement Programs
Introduction
Chapter 1 compared employer-provided retirement benefits conferred to public sector employees
in California with the benefits provided to federal and private sector workers. It also considered
the potential impact of two CFFR alternatives that would reform the benefits to be earned by
current or future California public sector employees via their future service. Chapter 2 considered
the broader context of total employee compensation.
In general, in Chapter 1 we found that for full-career employees, California governments provide
retirement benefits that are larger than what the federal government provides, and substantially
larger than what is available within the private sector. However, employees who do not remain
within a single retirement system for an entire career can sometimes do better under the federal
or private sector programs. We also found that retirement benefits for teachers can fall short of
what is provided to the other California public sector workers we considered; nevertheless, for
those who remain within CalSTRS for a career, benefits are still more generous than private
sector levels.
The two CFFR reforms would significantly reduce benefits to be earned through future service by
the full-career California public sector employee (a key exception being teachers who take early
retirement), although they would leave benefits well above private sector levels. The alternatives,
being more age neutral, would increase benefits for those who work in the California public
sector only at the earlier stages of their career.
Recap of CFFR Alternatives
Alternative A is based on a CFFR proposal that provides for a pension plan that is no more
generous than the Federal Employees Retirement System (FERS) plan, and a competitive
defined contribution plan. We modeled this alternative based on the current retirement income
provisions that apply to federal employees hired within the last 30 years, including the Thrift
Savings Plan (defined contribution) component. The federal plans pension provisions are
modified to include a wage cap, and a requirement that employees pay half of the expected costs
for future service; the defined contribution component was modified to include an additional
employer contribution with respect to pay above the pension wage cap. This alternative also
includes a Social Security replacement provision for employees not currently covered (including
teachers and CHP.) For the state employees, Alternative A eliminates retiree healthcare
dependent coverage after age 65, introduces an employee pre-funding requirement for the
retirees own post-65 coverage, and reduces employer subsidies with respect to pre-65 retiree
healthcare coverage, especially for those retiring before age 62.
Alternative B requires that employers pay no more than 6 percent annual costs for future service (9 percent
for safety employees) for non-health retirement benefits, and does not modify retiree health benefits. We
modeled Alternative B as a defined contribution plan with a dollar-for-dollar employer matching
arrangement up to the full 6 percent or 9 percent level. This alternative also has a defined benefit Social
Security replacement provision for employees not currently covered by that federal program.
More detail on the provisions of the current programs and Alternatives A and B is provided in
Chapter 1, including its appendix. The appendix to this Chapter includes information about the
assumed employee contribution rate under the pension components of Alternatives A and B.
In some cases, these rates differ from what was assumed in Chapter 1.
!

!

108 Chapter 3: Government Cost of Retirement Programs
Focus of This Chapter Employer Costs
Chapter 1 focused on the value of benefits received by sample individuals with alternative pay levels,
ages at hire and ages at termination. In this Chapter, we estimate the impact that the CFFR
reforms would have on costs to California government agencies to provide retirement benefits.
Naturally, cost depends on plan design features, such as benefit formulas for defined benefit plans
and matching provisions for defined contribution plans. For defined benefit plans, in particular,
costs also depend on demographic characteristics of the covered population, including age, service
years, and salary profile. And it depends on how events will unfold in the future:
How will employee pay levels change? Pensions are based on the employees highest salary
(averaged over one or more years).
When will employees terminate service, and at what ages will they start to draw pension
income and receive retiree healthcare benefits? These assumptions affect both expected
benefit levels and the expected number of years the retiree receives the benefits.
What level of inflation will retirees and their beneficiaries experience (since most public
systems adjust benefits for inflation), what medical costs will they incur, and how long will
they live?
For those defined benefit programs that are pre-funded, such as pensions, costs are also highly
dependent on the return to be earned on invested assets: the greater the earnings, the less that
must be set aside today to cover future benefit payments.
Actuarial valuations assign costs to defined benefit programs, based in part on assumptions about
those future unknowns. (The ultimate costs, of course, depend not on actuarial valuations, but on
how these many unknowns actually play out over time.) Where the CFFR reform alternative
involves a defined benefit program, the objective in this chapter is to estimate how employer costs
would be affected per the actuarial process currently used by the relevant California governing body. That is, we
do not make independent assessments of how the contingencies will play out, but instead simulate
how the valuation assumptions and methodology already in place would reflect proposed design
changes.
However, for one of the key unknowns the assumed rate of return on investments (which, for
state and local public pension funds, is the annual discount rate used to determine the present
values of future benefit payments) we also show results based on alternatives to the 7! percent
rate that is now used in pension valuations for each of the California public sector groups we
consider. We show the impact of instead using 6! percent and 5! percent.
1
The variability of
results based on alternative assumptions helps to illustrate the sensitivity of defined benefit costs to
unknown future outcomes.
As revealed in these results, a feature of the reforms that is as important as their impact on
expected costs is that they reduce governments exposure to cost surprises that arise when
things do not turn out as expected. By relying on defined contribution programs for some of the
benefits, they share the risk associated with investment return, inflation and longevity with
employees. Our illustration covers the first element of this risk sharing.

1
By law, a private sector employer must determine its required pension contributions using indexed rates
that have been close to 5! percent in recent years regardless of the investment return it expects its
pension fund to earn. The United States now uses 5! percent to determine its cost for federal employee
pensions, reflecting the investment policy used.


Key Steps In Determining Expected Costs
As noted previously, our models for defined benefit programs generally employ the actuarial
assumptions used in the respective official valuations, as disclosed in the most recent formal
report
2
or as separately provided. Detail regarding our approach, along with the key
assumptions, is provided in the appendix to this Chapter. Figure 1 outlines the steps in the
actuarial process used by the California governing bodies to set current year employer costs for
defined benefit plans.
Figure 1

Actuarial valuations generate projections of future benefit payments, allocate those amounts to
past and future service, and determine the unfunded portion of past service liabilities (if any).
This process results in assigning a current annual cost to the program: the sum of (1) a normal cost,
which is the value of benefits attributable to an additional year of service, and (2) an amortization
cost, which is the current years installment toward the difference between the value of benefits
attributable to past service and accumulated assets. This total cost is also referred to as the annual
required contribution, or ARC.
As noted in Chapter 1, state and local governments do not pre-fund retiree health benefits, but
instead finance them on a pay-as-you-go basis. However, a directive issued by the Governmental
Accounting Standards Board in 2004 requires agencies to record and report the annual costs of
retiree health care on an actuarial basis similar to that used for pensions. Given the lack of pre-
funding, there are no assets available to offset the actuarial accrued liability shown in Figure 1.
Hence virtually all of the accrued liability is unfunded.
3

2
As of the writing of this Chapter, the most recently published report for state employee pensions is as of
June 30, 2009, and the most recent reports for state employee retiree healthcare benefits and for teachers
pensions is as of June 30, 2010.
3
The June 30, 2010 retiree healthcare actuarial valuation covering all current and retired employees of the
State of California and their beneficiaries determined that the normal cost for 2011-12 is about $2.2 billion,
and that the 2011-12 installment to amortize the unfunded accrued liability over 30 years is almost $2.5
!

!

Defined Contribution Programs. Compared to defined benefit programs, costs for defined
contribution programs are determined in a relatively straightforward manner. Because the
employers commitment is limited to the contribution to be credited for the year and does not
extend to the benefit amounts to be paid in the future, it is not necessary to undertake actuarial
valuations, or to estimate normal costs and amortization costs each year. Rather, the
employers contribution for a year is simply based on the plan design, employee salary levels and,
in the case where there is an employer match, employee contribution levels.
Pension Amortization Costs Currently Large and Rising
Weaker than expected investment returns, unanticipated changes in demographic factors (such as
lengthening life spans), and larger than expected pay increases in past years have resulted in
major unfunded liabilities for state and local government pension funds. Even after accounting for
some above-expected investment returns in 2010 and early 2011, and even using the funds own
assumptions about future investment return, the average of these shortfalls is likely more than 25
percent of the funds accrued liabilities (based on market value of assets). These unfunded
liabilities have already led to substantial increases in the component of employer contributions
dedicated to amortizing the shortfalls. As shown in Figure 2, more than one-half of the total
employer pension rates for state non-safety and CHP employees relate to amortization of
unfunded past service liabilities. We expect this component to rise further in the near term: under
the actuarial smoothing methods that CalPERS and other pension funds employ to defer
recognition of unexpected investment gains and losses, thus far only a portion of the investment
shortfalls experienced in 2008-9 have been incorporated into the funds rate-setting calculations.
A key point here is that, even with pension reform, state and local funds will likely continue to
face rising pension costs related to deferred effects of past investment results.
Figure 2
Employer Pension Costs in 2011-12 as Percent of Payroll

billion, for a total cost for that year of $4.7 billion. Because these total annual costs are not being met via
pre-funding, the unfunded accrued liability will grow each year with interest and the unpaid normal cost,
leading to a pyramiding of amortization costs over the future.



A Note About Actuarial Costs and Budget Outlays
Our estimates focus on the impact of CFFR proposed reforms on actuarially determined employer costs
what state and local governments should be setting aside each year to cover the future pension
and retiree health care benefits that their employees are earning. These measures provide a good
picture of the ultimate financial impacts of the reforms. However, they do not always provide a
good indication of how reforms will affect cash (budget) outlays in a particular year.
There is a timely effect on budget outlays for defined contribution programs, as well as for defined
benefit programs where the policy is to fully fund the ARC. However, for defined benefit
programs where the policy is not to fully fund the ARC, the effect of reforms on budget outlays
can be delayed for years even decades.
One example is CalSTRS, where pension contributions are set in statute, and are currently at
levels that are insufficient to cover the ARC. Unless statute changes lead to full current funding of
the ARC, reforms that reduce teacher pension costs will have only a delayed budget impact.
A more extreme example is retiree health care. Given the pay-as-you-go policy, a reform can
reduce current annual cost by a significant margin but have only a very modest effect on near-
term budget outlays. Under the pay-as-you-go policy, the full effects will not show up in cash flow
until after the affected employees retire.
This difference can lead to near-term budget outcomes that are counter to the long-term change
in costs. For example, if a reform lowers retiree health costs substantially and raises combined
pension and thrift program contributions modestly, the long-term result is a significant net cost
saving. However, given the delayed cash effect for the health care provisions, the near term
impact on the budget may be temporarily higher outlays.
Specific Estimates
In this section, we develop specific employer cost estimates for the reform proposals as they would
apply to (1) State Miscellaneous (non-safety) employees, (2) California Highway Patrol employees,
(3) public school teachers, and (4) a hypothetical local non-safety employee group. For the local
employee group we consider the impact for three variations on a current pension design available
to local agencies contracting with CalPERS.
In the subsequent section, we use these results to make more generalized inferences about the
impacts of the CFFR alternatives on state and local governments in California.
!

!


Timelines
Our comparisons provide two snapshots of how the CFFR reforms would affect employer costs.
We show the near-term impact in the scenario where the reforms apply to future service of
existing employees, and the ultimate effects of the reforms, which is the impact at the point in the
future when they apply to the full careers of all employees in the systems.
The figure below how these cost changes would emerge over time, using as an illustration the
pension changes in Alternative A. It shows that, if the reform applies to future accruals of existing
employees as well as new hires, it will result in an immediate decline in the normal cost. However,
if the reforms apply only to new workers hired after the effective dates, the reductions will phase
in over many years, as the workforce turns over. (Though for some pension systems particularly
at the local level the requirement that employees start paying one-half of normal costs would
result in partial savings immediately.) The figure also shows a modest cost decline over time under
current law, as the effects of the recently enacted benefit reductions for new hires apply to more
employees as the workforce turns over.
Timeline of Alternative A Pension Changes. CalPERS State Non-Safety Employees
(Normal Cost as Percent of Payroll)

Similar timing issues apply to retiree healthcare changes included under Alternative A: the
reduction in actuarial costs will be experienced much more rapidly if the change applies to current
employees, rather than just to future hires. However, since California finances retiree health care on
a pay-as-you-go-basis, the reduction in costs will not translate into budget savings until the
employees retire and start drawing the benefit. If the retiree healthcare reforms apply only to future
hires, the cash effects would not show up until those not-yet-hired employees themselves retire;
under this scenario cash savings over the next 20 years would be negligible.
The timeline for Alternative B depends on how it is implemented. We modeled this alternative as a
defined contribution program that could be implemented for future accruals of existing employees
or just for new hires. Under these scenarios, the timelines for the programmatic changes would be
similar to that shown above for Alternative A. Even in the case where the change to a defined
contribution program applies only new employees, however, this alternative could result in
immediate savings, if employer normal costs for existing pension systems were also limited to
6 percent of payroll (9 percent for safety employees).


For each employee group, we show up to three sets of comparisons:
The immediate impact of the alternatives, under the scenario where the changes apply to
future accruals of existing employees.
The ultimate, or steady-state, impact of the reforms relative to current law that is, the
relative cost levels once the reforms have applied for the full careers of all employees in
the system (this can be more than 35 years in the future).
Revised estimates of the ultimate impacts for the retirement income programs using
alternative assumptions about the funds annual rate of return on invested pension assets
a very important issue when considering the risks that unanticipated outcomes can
pose to costs of retirement systems.
Behavioral Effects. Our comparisons are based on static estimates of how the reforms would
apply. That is, they reflect application of the alternative systems to the existing cohort of state
employees, using the funds current actuarial assumptions about wage growth, turnover, retirement
rates, and other factors. Because of changes in hiring practices or other socio-economic factors,
the employee population of the future may have a different demographic profile than todays
cohort. And future employee behavior may vary from whats predicted under the current
actuarial assumptions whether due to revised behavioral incentives within the reforms or
otherwise. Given our approach, the cost estimates here will not reflect the impact of those
differences.
Despite the exclusion of these behavioral effects, we believe that static estimates provide a
reasonable indication of the general cost impacts of the reforms. This is particularly the case in
the near term. If the changes apply to existing employees, the pension and defined contribution
reforms are unlikely to trigger behavioral change for some time, since in general pensions earned
prior to the reform effective date will outweigh benefits earned thereafter.
4
In the long run, where
all benefits will have been earned under the reform provisions, things are more uncertain. For
example, provisions within the pension component of Alternative A that improve benefits upon
reaching certain age and service thresholds may lead to retirement patterns that differ from what
is currently expected. This could mean larger pension costs under that Alternative than we
estimate here using current retirement assumptions, but also smaller retiree healthcare costs.
In general, more caution is appropriate when looking further into the future.
CALPERS Miscellaneous (Non-Safety) Employees
Our first comparison covers Californias main non-safety category, having 159,000 active
members. The accrued pension liability for this group, including retirees and other inactive
members, was $75 billion as of the June 30, 2009 valuation (reflecting the 7! percent investment
return assumption.) Employees first hired through 2010 are subject to the 2 percent at age 55
pension formula, based on 12-month average pay for those hired before 2007 and a 36-month
average for others. Under legislation enacted in 2010, employees first hired after January 13, 2011
are covered by a less generous 2 percent at age 60 formula. Employee contributions were
5 percent of pay in excess of $513 per month prior to November 2010, but were increased to
8 percent of such excess pay for later periods.
Based on the most recent valuation, current non-safety employees account for about $19 billion of
Californias $60 billion accrued liability for retiree health care benefits earned by all active and

4
Behavioral pressure exerted by the retiree healthcare reforms under Alternative A could have a more
immediate impact and delay some retirements; on balance, such delays would tend to increase the savings
under that proposed reform beyond what is estimated here.

!

!

retired state employees through June 30, 2010. (Current retirees in this group account for an
additional $18 billion.) The ARC for 2011-12, including the cost for benefits to be earned in that
year as well as that years installment to amortize unfunded accrued liability, is $2.8 billion.

Immediate Impact Assuming Changes Apply To Existing Employees
As shown in Figure 3, the combined ARC for pension and retiree health benefits under existing
law totals 46 percent of pay, consisting of 18 percent for pensions and another 28 percent for
retiree health. Under the scenario where the reforms apply to existing employees, combined costs
would drop to 34 percent under Alternative A, mainly due to reductions in the retiree health care
component. On the non-health side, Alternative A would reduce costs by slightly less than
1 percent, reflecting a nearly one-third decline in the ARC for the pension but an offsetting cost
increase for the new defined contribution program. Employer costs under the Alternative B
defined contribution program would be almost 3 percent lower than current law pension. (As will
be discussed below, these cost comparisons for non-healthcare benefits look different under
alternative assumptions about future pension investment return.)
Figure 3
CalPERS Non-Safety Immediate Change in Annual Cost if Alternatives Apply to
Existing Employees\a

a\ Includes the ARC for pensions and retiree health care defined benefit programs, plus annual employer
payments for defined contribution programs.

Ultimate Impact of Alternatives
Figure 4 shows a similar set of comparisons as Figure 3, except that it looks much further down
the road, when the reforms will have applied to the full careers of all employees. Accordingly, our
current law pension component here reflects the recently enacted CalPERS 2 percent at age 60
benefit formula for new hires, which is less generous than the provisions that apply to the current
workforce. As a result, the long-term impact from the pension reforms is reduced compared with
the short-term impact.
The cost impact for interim years is discussed in the nearby box.
Estimates of ultimate costs do not include amortization payments. In order to focus
on the costs that result from additional employee service under the current or alternative
programs, we are excluding any amortization-related costs that may still exist once these
2% at age 55


alternatives are fully phased in. The level of amortization cost at that future time will depend on
events during the intervening years, including whether employer contributions will be made that
fully cover the ARC, and whether investment returns and other experience will conform to
current actuarial assumptions.
As a consequence, our ultimate cost comparisons include only normal costs for the pension and
health care components, and the annual employer payments under the defined contribution
program.
Figure 4
CalPERS Non-Safety: Ultimate Effect on Costs
Excludes Amortization Costs\a

a\ Includes only normal costs for pensions and retiree health, plus annual employer payments to the
defined contribution programs.

Both reform measures result in cost savings relative to current law, with all of the savings under
Alternative A coming from the retiree health reforms (using the current 7! percent pension
investment return assumption). Compared to benefits scheduled under current law, combined
pension and defined contribution costs are slightly higher under Alternative A. The cost of the
pure defined contribution program under Alternative B is modestly lower than current law.

Two factors account for the relatively modest long-term effects of the non-health components of
the reforms. First, normal costs for the pension under existing law will be lower in the future than
today, reflecting application of the recently enacted reforms (in pay averaging period and in
benefit formula) to the future workforce, and full-career application of increased employee
contribution rates. Second, normal costs (relative to the benefits received) are lower for this group
than others we examine. This is partly due to this groups relatively higher degree of expected
mobility. Per the assumptions used by CalPERS actuaries, these employees are less likely to
remain in service until retirement age than are other employees we consider, such as teachers.
This limits defined benefit costs under the existing system and potential savings under the
reforms.

It also should be noted that our estimates for the alternatives assume full implementation of the
maximum defined contribution designs. Under Alternative A, this leads to an expected employer
cost of about 5 percent of payroll for the current federal thrift design (including the additional
2% at age 60
!

!

contribution based on pay in excess of the cap used under the pension component). The CFFR
proposal does not require adoption of the federal thrift design. Rather, it authorizes agencies to
offer a competitive defined contribution plan. When combined with the pension component, a
competitive defined contribution plan could feature less generous employer contributions than the
current federal thrift savings plan, with lower costs.
Effect of Alternative Investment Return Assumptions on Pension Costs
The pension comparisons above are based on the CalPERS discount rate of 7.75 percent, which
reflects their expected long-term rate of investment return. The Figure 4 results are replicated in
the left column of Figure 5 below. The results for Alternative A include the combined total of the
pension and defined contribution components of the modified federal plan, and Alternative B
reflects the cost of the pure defined contribution plan. Retiree healthcare costs are excluded from
this comparison.
The discount rate assumption has a significant impact on current pension costs. What is the
appropriate rate for this purpose? State and local government pension funds generally use their
expected long-term rate of return on investments for the discount factor. These rates are usually
around 7.5 percent to 8 percent, which is near the funds long-term average returns for certain
historical periods. However, many observers consider these rates to be inappropriate given the
uncertain nature of future investment returns, and the fact that benefits are fully guaranteed.
If the discount rate is reduced, the result is higher current costs.
This issue does not arise for defined contribution plans, since the employers cost is not affected by
future investment returns. (Investment risk is borne by the employee.) Future investment return
matters much more under current law programs, which rely exclusively on pension plans, than
under the proposed reforms, which rely partly or fully on defined contribution plans.
As indicated in Figure 5, lowering the assumed investment return has a smaller impact on
employer costs under the alternatives, making them increasingly less expensive relative to current
law. Whereas annual costs under current law and the two alternatives are roughly similar based
on the 7.75 percent assumption currently used by CalPERS, things look strikingly different using
alternative assumptions.
If the discount rate were lowered to reflect a more conservative investment return assumption, the
current annual cost would immediately adjust to reflect this. If the assumed rate is not lowered but
actual investment returns end up being consistent with the lower rates shown in Figure 5, the state
would experience steady cost increases over future periods as lower-than-expected investment
returns add to unfunded past service liability and, hence, amortization costs; in this scenario, the
cost results in Figure 5 reflect the resulting average over the long-term.
Regardless of the mechanism, Figure 5 illustrates one of the main benefits of the CFFR reforms.
They shield, to varying degrees, state and local employers from the major cost increases that can
occur when investment or other results turn out less favorably than expected. They lessen the
chances that current fiscal hardships created by past shortfalls will be repeated in the future.


Figure 5
Sensitivity of Annual Employer Costs to Investment Return Assumption
(Costs as a percent of payroll)
Assumed Rate of
Return
Annual Costs To Fund Retirement Income Programs
Current Law Alternative A Alternative B
7.75 percent 7.0 7.7 5.4
6.75 percent 10.4 9.1 5.4
5.75 percent 14.8 11.0 5.4
a\ Ultimate effects of alternatives. Includes normal costs for pensions plus annual employer contribution to
defined contribution programs under the alternatives. Does not include retiree health care costs. Differing
assumptions about investment returns would also change amortization costs, but in an identical fashion
among the alternatives.
CalPERS CHP Employees
Our second set of comparisons covers the CalPERS CHP category. This group includes about
7,400 active members, with accrued pension liabilities of $7.3 billion as of June 30, 2009,
including $4.4 billion for retired officers. For those hired before 2011, pensions are earned under
an enhanced safety formula that provides 3 percent of highest 12-month salary per year of service
(maximum 90 percent) beginning as early as age 50; those hired after 2010 who retire before age
55 receive reduced amounts. CHP officers do not participate in Social Security.
In past years, the employee pension contribution rate was set at 8 percent, though collective
bargaining agreements included employer pickup of some or all of the employees share. The rate
beginning in mid-2010 is 10 percent of pay in excess of $863 per month.
Based on the most recent valuation of retiree health costs, active employees in this group account
for about $1 billion of the State of Californias $60 billion accrued liability for health care benefits
owed to active and retired state employees. (Current retirees in this group account for an
additional $2 billion in liabilities.) The annual cost for benefits expected to be earned in 2011-12
is $103 million, and the installment towards amortizing unfunded past service liability for that
year is $132 million.
Immediate Impact Assuming Changes Apply To Existing Employees
Figure 6 shows that, under current law, the ARC for the pension and retiree healthcare defined
benefit programs combined is 62 percent of payroll, split about evenly between the two
components. Under Alternative A, combined costs for the pension, retiree health, and new
defined contribution programs would equal 47 percent of payroll. Under Alternative B the
combined total would be about 56 percent of payroll.
!

!

Figure 6
CalPERS CHP
Immediate Change in Annual Cost if Alternatives Apply to Existing Employees \a

a\ Includes the ARC for pensions and retiree health care, plus annual employer payments for defined
contribution programs.
The reform proposals would result in significant reductions in non-health retirement costs. These
costs would drop from 31 percent under current law to about 25 percent under each alternative.
The pension totals for both alternatives include the Social Security replacement provisions that
apply to this group.
A significant factor reducing pension costs under Alternative A is the wage cap, which would
apply to a larger share of employees in this group than, for example, the non-safety or teachers
categories. In addition, Alternative A reforms would significantly lower current retiree health
care costs.
Ultimate Impact of Alternatives
Figure 7 shows the ultimate impact of the reforms on annual employer costs, when all employees
are fully covered by the alternative programs. The long-term comparisons do not include
amortization costs. Also, the current-law pension component for this comparison is the
3 percent at age 55 benefit formula that applies to employees hired after mid-2010. The figure
shows that the long-term savings are similar to the immediate savings shown above, though the
pension related component is slightly smaller. This is mainly due to the above-mentioned benefit
formula reduction, which will lower future normal costs for the pension under current law.

3% at age 50


Figure 7
CalPERS CHP: Ultimate Impact on Costs

a\ Includes only normal costs for pensions and retiree health programs, plus annual employer payments to
the defined contribution programs.
The sensitivity of employer pension costs to investment return is greater for this group than for
the non-safety employees, reflecting, in part, their younger ages, and that a larger portion of the
total benefit is funded via employer contributions. This difference in investment return sensitivity
has significant implications for the state, and even greater implications for those local agencies
where a large proportion of annual budgets are for safety employee salaries and pensions.
Figure 8
CalPERS CHP
Excludes Amortization Costs \a
Assumed Rate of
Return
7.75 percent 14.0 11.2 10.0
6.75 percent 21.2 14.2 12.0
5.75 percent 30.6 18.2 14.7
defined contribution programs under the alternatives. Does not include retiree health care costs. Differing
assumptions about investment returns would also change amortization costs, but in an identical fashion
among the alternatives.
CalSTRS (Teachers)
CalSTRS is a statewide fund that provides pension benefits to certificated teachers and
administrators. It has 440,000 active members, and an accrued liability of $196 billion as of
June 30, 2010 for all 850,000 active and inactive members. Employees receive benefits under a
3% at age 55
!

!

formula that provides a percentage of the highest 36-month average salary (12-month average
with 25 years of service) for each year of service that varies from 1.1 percent at age 50 to
2.4 percent at age 63. Teachers contribute 8 percent of pay (6 percent from 2001 through 2010),
and do not participate in Social Security. Contributions are set by law at a level that does not
currently cover the ARC. For 2010-11, the ARC is 25.5 percent of payroll, but combined school
district and state contributions to CalSTRS cover only 11.3 percent. This annual shortfall adds to
a growing unfunded liability.
Pension costs for teachers are higher (relative to the benefit levels) than for the state non-safety
group. This reflects various factors, including that a higher proportion of teachers are expected to
remain in the system until retirement years.
Retiree health benefits are provided on a district-by-district basis, and, in general, are less
generous than those provided to state employees. Although adoption of the Alternative A retiree
health reforms could result in savings for some districts, the impact would vary greatly by district
and is not modeled here.
Effects of Alternatives
Figure 9 shows the immediate effect that application of the reforms would have if applied to
future service accruals of existing employees as well as to new hires. It shows that under current
law, the ARC is about 25 percent of pay. Application of Alternative A would result in little change
to that total, and Alternative B would result in a moderate reduction.
Figure 10 shows that a generally similar picture emerges for the long-term comparisons. Under
Alternative A, the normal cost of the pension program would fall by over one-third. This decline
would be offset, however, by added costs from the new defined contribution program. As a result,
the combined cost for the pension (including the Social Security replacement) and defined
contribution programs is slightly higher than under current law. The Alternative B defined
contribution and Social Security replacement provisions would cost about 4 percent less than the
normal cost for the pension program under current law.
Figure 9
CalSTRS: Pension and Defined Contribution Programs Immediate Change in
Costs if Alternatives Apply to Existing Employees \a

a\ Includes the ARC for pensions plus annual employer payments for defined contribution programs.



Figure 10
CalSTRS: Ultimate Impact on Cost

a\ Includes only normal costs for pensions and annual employer payments to the defined contribution
programs.
Figure 11 shows the impact of alternative pension investment return assumptions on annual costs.
Again, costs under the reform alternatives are less sensitive to investment earnings assumptions
than are current law costs. Assuming a 5.75 percent rate of return, Alternative A would produce a
savings of over 4 percent of pay relative to the current law baseline and Alternative B would result
in a savings of over 13 percent of pay.
Figure 11
CalSTRS (Teachers)
Assumed Rate of
Return
7.75 percent 11.3 11.9 6.6
6.75 percent 16.5 14.9 8.0
5.75 percent 23.1 18.7 9.9
defined contribution programs under the alternatives. Differing assumptions about investment returns
would also change amortization costs, but in an identical fashion among the alternatives.
!

!

Representative Local Government Funds
There are over 1,500 local agencies operating (or contracting for operation of) pension systems in
California. The funds have a variety of benefit formulas, employee contribution requirements,
and other features. Within pension systems, there are also multiple tiers of benefit formulas that
apply to employees hired at different times. For this reason, the CFFR alternatives would have a
wide range of impacts on local agency costs, and no one system is truly representative of these
impacts.
However, our review of local funds indicates that, on average, local pension plan designs are more
generous than the state in terms of benefit formulas, the determination of final compensation, and
required employee contributions. Moreover, a majority of local governments have provisions in
collective bargaining contracts calling for employer pickup of part or all of the employees shares
of pension contributions. Though recent labor contract negotiations have resulted in some scaling
back of this practice, it remains widespread at the local level.
5
A related practice is employer
contributions to supplemental defined contribution plans.
To show a reasonable range of effects that the CFFR alternatives would have on local employer
costs, we have developed detailed estimate for three variations of benefit designs for local non-
safety employees, taking into account the features described above:
A 2 percent at age 55 system, with final compensation determined by highest annual
salary, 2 percent annual inflation adjustment (in addition to 80 percent pension
protection allowance) and employee contributions equal to 7 percent of pay. Special
compensation is recognized, and is assumed to equal 5 percent of final years base pay.
A variation of design (1), where the employer picks up 100 percent of the employees
required contribution.
A further variation of design (1) that is less common, but still often used. The formula
provides a 2.5 percent at age 55 benefit, with nominal employee contributions of
8 percent of pay. We assume that the employer picks up 50 percent of this employee
contribution.
Our estimates show the ultimate impacts when the alternatives apply to full careers of all
employees. As with our previous comparisons of ultimate impacts, these figures exclude
amortization costs. They include just normal costs for pensions and the new defined contribution
components of the two alternatives.
To model costs for these representative local systems, we assumed that its employee population
has the same demographic characteristics age, service, salary and gender distribution as the
state non-safety population as of June 30, 2009.
Consistent with Chapter 1, we did not model the effects of the Alternative A retiree health
reforms on our representative local agencies. Benefits provided by local agencies vary from
agency to agency, but are, on average, less generous than what is provided by the state.
Results
Figures 12 through 14 show that the CFFRs proposed reforms would significantly reduce costs
relative to the plan designs outlined above. Figure 12 shows that, for an agency offering a

5
For example, a 2010 study found that of 17 cities in San Diego County that contracted with CalPERS for
administration of their pensions, 13 picked up some or all of the non-safety employees required
contributions. For safety employees, 14 out of 17 agencies picked up some or all of the employees
contributions. (See Phase I Update, San Diego Pension Plans. San Diego Taxpayers Association,
September 20, 2010).


2 percent at age 55 benefit structure with no employer pickup of employee contributions,
Alternative A would reduce these costs by almost one-quarter. Alternative B would lower costs by
nearly one-half.
Figure 12
Local Government Design (1): 2 percent at 55, No Employer Pickup.
Ultimate Impact, Excludes Amortization Costs\a

a\ Includes only normal costs for pensions, plus annual employer payments to the defined contribution
programs.
Figure 13 shows results for design (2). For an agency offering this program, the reduction in
annual costs from the CFFR pension reforms would be substantial over 8 percent of payroll
under Alternative A and over 10 percent under Alternative B.
Figure 13
Local Government Design (2): 2 percent at 55,
Employer Picks Up Employee's Contributions

programs.
!

!

Figure 14 shows results for design (3). For an agency offering this program, the eventual savings
from the CFFR pension reforms would be similar to the previous set of comparisons almost
8 percent of payroll under Alternative A and almost 10 percent under Alternative B.
Figure 14
Local Government Design (3): 2.5 percent at 55
Employer Picks Up One-Half of Employees Contribution

programs.
Together, figures 12 through 14 indicate that the CFFR alternatives would have particularly
significant impacts on local plans involving richer benefit formulas and/or employer pickup of
some or all of employee contributions.
Figure 15 shows the impact of alternative pension investment return assumptions on the costs for
the representative local pension system using design (3). For richer local plans, the difference in
employer costs becomes dramatic with lower investment return assumptions. If the assumed rate
is reduced to 5.75 percent, for example, the savings under Alternative A relative to current law
would be almost 16 percent of pay and under Alternative B would be over 20 percent of pay.


Figure 15
Local Government Design (3): 2.5 Percent at Age 55 System
Amortization Costs Excluded \a
Assumed Rate of
Return
7.75 percent 15.1 7.4 5.3
6.75 percent 19.7 8.5 5.3
5.75 percent 25.8 10.1 5.3
defined benefits programs under alternatives. Differing assumptions about investment returns would also
change amortization costs, but in an identical fashion among the alternatives.
Effects of CFFR Alternatives On Statewide Costs
Determining the precise impacts of the CFFR alternatives on state and local governments would
require detailed valuations of over 2,000 local agency plans. While such detailed analyses are
beyond the scope of this report, it is nevertheless possible to tentatively draw some general
conclusions from our modeling efforts.
Current Employer Costs
Figure 16 provides background on statewide pension costs. It shows that in 2008-09 (the most
recent year for which statewide data is available), 2.4 million public employees were members of
pension systems in California, the funds had total accrued liabilities of $682 billion, and total
contributions to the funds were $25 billion. Of the total contributions, slightly over $14 billion
were from employers, $9 billion were from employees, and $1 billion were from other sources
(mostly state General Fund contributions to CalSTRS). (The employee contributions include an
unknown but potentially significant amount of contribution picked up the by the employer.)
Both employer and employee contributions have increased significantly since 2008-09, mainly
reflecting: (1) gradual recognition of the 2008-09 investment losses, and (2) changes in statutes and
collective bargaining agreements affecting employee contributions. We estimate that total
contributions in 2011-12 will be about $28 billion, including about $16 billion in employer
contributions. These contributions will likely rise further in the near term future as the effects of
the 2008-09 losses continue to be phased in.
Aggregate statewide information is not available for retiree health coverage. However, we know
that, in addition to Californias $60 billion unfunded liability, UC has an unfunded liability of
$14 billion, the county of Los Angeles has an unfunded liability of $24 billion, and numerous
other local agencies have aggregate liabilities in the hundreds of millions to billions of dollars
range. Overall, statewide commitments for future retiree health care are well over $100 billion,
and normal costs to cover just annual benefit accruals are around $5 billion statewide. Costs to
amortize unfunded liabilities from past service would be an additional $5 billion.
!

!

Figure 16
Characteristics of State and Local Pension Funds In California
2008-09
Active
Members
Actuarial Accrued
Liability
Annual Contributions in 2008-09
(billions)
Pension Fund: (thousands) (billions) Employer Employee Other
Statewide\a 1,868 $498 $9.7 $7.3 $1.3
Counties 256 113 3.1 1.2
Cities 105 66 1.1 0.6
Districts 100 5 0.3 0.1
Total 2,329 $682 $14.2 $9.2 $1.3
Source: Public Retirement Systems Annual Report, Fiscal Year 2008-09. State Controllers Office.
a\The statewide totals in this table include local agencies contracting with CalPERS.

Statewide Effects of CFFR Proposals

Adoption of the CFFR proposals would result in savings at both the state and local level. The
estimates below show the ultimate impacts of the changes. If the reforms were to effect future
accruals of existing employees, the pension impacts would be immediate (and for state programs,
modestly larger than shown below)
6
.

However, if the program reforms apply only to newly hired employees, the timing of the impacts
is more complicated.

In the case of Alternative A, the cost changes associated with the programmatic reforms
would emerge slowly as the workforce turns over. However, the requirement that
employees pay one-half of normal costs for pensions would have an immediate impact.
The savings from this provision would be modest at the state level, since most employee
groups are already paying close to one-half of normal costs. The savings would be much
larger at the local level, however, since many employees are currently paying significantly
less than one-half of normal costs, due in part to employers pickup of their contributions.
We estimate that this provision would reduce local employer costs by well over $2 billion
annually.

As we modeled Alternative B (a defined contribution program), under the scenario where
the reforms only apply to new employees, the cost savings from the shift to the new
system would also emerge slowly over time as the workforce turns over. However, if the
caps on employer normal costs (6 percent of payroll for non-safety and 9 percent for
safety employees) also apply to existing pension systems, most of the savings we identify
below would occur immediately.

6
As noted in Chapter 1, our analysis looks solely at the financial impacts the proposed alternatives under
different implementation scenarios. We have not determined whether the provisions or implementation
scenarios comply with constitutional provisions, labor law, or IRS regulations.


Effects on State-Level Funds

Using the funds actuarial assumptions about investment returns, both of the CFFR alternatives
would result in cost reductions for state-level funds (CalPERS State and Schools, CalSTRS, UC,
Judges, and Legislators).

The pension and thrift provisions of Alternative A would reduce annual costs by around
0.25 percent of payroll, or the low hundreds of millions annually in todays dollars.
The Alternative A retiree health reforms would reduce retiree health care costs by about
$2 billion in todays dollars.
Alternative B would result in an annual reduction in employer costs of about 4 percent
($2.7 billion in todays dollars) for pension and defined contribution programs, but would
have no impact on retiree health benefit cost accruals.
Estimates Assume Full Adoption of Federal Thrift Plan. Alternative A assumes the full
implementation of the federal thrift plan, with an expected employer costs about 5 percent of
payroll. If the state were to offer a less-generous plan say a 3 percent match the modest
reduction noted above would become significantly larger. (Each 1 percent decline in annual
contributions translates into about $700 million in state-level savings.)
Impact of Lower Investment Returns On Relative Costs
While the initial costs assigned to the alternatives would be based on the actuarial assumptions
used by the respective pension systems, the ultimate cost will depend on actual experience relating
to investment returns and other factors. As noted in our estimates for specific funds, if the rate of
return is less than currently assumed by state and local pension funds, then both current law and
the alternative systems will experience cost increases. However, the extent of the cost increases
will be significantly less under the alternatives, because of their partial or full reliance on defined
contribution programs.
For example, under Alternative A, the modest reduction in state costs cited above using a
7.75 percent investment rate of return assumption would expand to over $3 billion (5 percent of
payroll) using a 5.75 percent return assumption. Under Alternative B, the $2.7 billion savings
under the 7.75 percent rate-of-return assumption is about $7.5 billion (11 percent of payroll)
under a 5.75 percent rate.
Of course, if pension investment returns exceed the assumed 7.75 percent over the long-term
future, the opposite relation holds.
Effects on Local Governments
We estimate that adoption of the CFFR alternatives would produce significantly larger overall
pension-related savings at the local level (which includes CalPERS contracting agencies as well as
county, city, and district funds). Based on our modeling of some of the more common benefit
structures, we believe that implementation of Alternative A would reduce average local costs by
5.5 percent to 7.5 percent of payroll or about $3 billion to $4 billion in todays dollars. The
Alternative B savings would be 7.5 percent to 9 percent of payroll, or $4 billion to $5 billion in
todays dollars.
The ranges partly reflect differing assumptions regarding the magnitude of employer pickup of
employee contributions and employer contributions to supplemental defined contribution
retirement accounts. As noted in our detailed discussion for local governments, the former
practice, in particular, is common at the local level, though the exact magnitude of the cost
impact is uncertain.
!

!

We did not explicitly model local results related to retiree health care, but it is likely that adoption
of CFFR retiree health reforms would produce an additional savings (potentially in the high
hundreds of millions of dollars annually in todays dollars).
As with the state, these comparisons are based on current investment rate of return assumptions
used by public pension systems. Use of a lower assumed rate of return produces much larger
savings at the local level as well.
Conclusion
As noted above, full implementation of both CFFR alternatives would reduce employer costs at
both the state and local level over time. Using the actuarial assumptions of the respective funds
about investment rate of return and other factors, the pension reform provisions of Alternative A
would result in modest declines in state pension related costs and more significant reductions in
retiree health care costs. Alternative B would produce moderate savings in state pension costs.
It does not reform retiree health care.
At the local level, we estimate that both alternatives would reduce overall employer costs for
retirement income programs by a significant amount, though effects would differ from one agency
to another depending on their existing benefit programs. The larger overall savings compared to
the state reflects relatively richer plan designs and the prevalence of employer pickup of employee
contributions at the local level.
In addition to savings with respect to expected costs, the proposals would reduce governments
exposure to unexpected costs arising from weaker-than-expected investment returns and other
factors. The reduction in costs and volatility will benefit the state in the long term, by reducing
funding pressures that retirement costs will impose on other areas of state and local budgets.
The reforms will also pay dividends in the near term, through their impacts on state and local
governments long-term liabilities, their credit outlooks, and, potentially, their access to
credit markets and thus their ability to finance infrastructure spending.
7

7
See for example, Combining Debt and Pension Liabilities of U.S. State Enhances Comparability. Moodys Investors
Service, Special Comment, January 26, 2011.

Chapter 3: Appendix 129

!
Chapter 3 Appendix
This appendix summarizes our approach to the employer cost comparisons for the defined benefit
and defined contribution programs discussed in Chapter 3.
General Approach
For defined benefit programs, including pension and retiree healthcare benefits, the respective
California governing bodies currently determine employer cost based on annual actuarial
valuations incorporating assumptions and methods that they select. For these programs, our
approach was to simulate the cost results that those valuations would determine for the proposed
reforms, generally using the same actuarial methods and assumptions as described in the most
recent valuation report. As of this Chapters preparation in June 2011, the most recent report is as
of June 30, 2010 for retiree healthcare benefits covering state employees and for pension benefits
for teachers, and as of June 30, 2009 for the other pension benefits considered.
For the defined contribution component of Alternatives A and B, we calculate the employer
contribution as a percent of payroll for a representative year (2013), based on the same population
modeling and salary increase assumptions as apply for defined benefit plans, using additional
assumptions regarding employee contribution rates under matching arrangements.
These processes are described further below.
Population Modeling
We modeled each active employee population by reference to the five-year age and service
summary included in the most recently published actuarial valuation report for that group. For
state non-safety and CHP employees, this included the number of employees within each five-
year band covering age and service (for example, number of employees between ages 35 and 39
with between 10 and 14 years of service), and their average pay on the valuation date. For
teachers, report information did not include average pay by age-service unit, so it was necessary
to impute it to each of the 81 populated age-service cells.
1
Average pays as of June 30, 2009 were
then increased to reflect the overall change in payroll as of June 30, 2010.
The central age and central service for a cell is treated as its age and service employees
between ages 35 and 39 with between 10 and 14 years of service are treated as age 37 with 12
years of service. Judgments are made as appropriate for border cells: for example, teachers
younger than age 25 with between one and five years of service are treated as age 24.
Each cell is then subdivided into two sub-cells that each inherit the age and service of the parent.
One contains the majority of the employees in the parent (85 percent for teachers, 90 percent for
state non-safety employees and 95 percent for CHP employees) with assigned pay equal to a
portion (98 percent for teachers, 90 percent for state non-safety employees and 99 percent for
CHP employees) of the average for the parent, and another containing the remaining employees.

1
This was done as follows: (1) average annual pay among teachers under age 25 with zero years of service
as of June 30, 2010 is assumed to be $37,000; (2) for service up to 20 years, average pay for a cell is assumed
to be 21.9 percent higher than for a cell with five fewer years of service; (3) among cells with a given level of
service, pay is assumed to be higher by 0.3 percent for each additional year of age. The resulting
distribution produces an average annual pay as of June 30, 2010 among all 441,544 active teachers that is
very close to the $59,507 average for the entire group shown in the valuation report.

130 Chapter 3: Appendix

For example, a cell containing 1,000 employees with an average annual pay of $70,000 would
divide into two sub-cells, as follows.
teachers: (i) 850 earning $68,600, and (ii) 150 earning $79,882
state non-safety employees: (i) 900 earning $63,000, and (ii) 100 earning $133,000
CHP employees: (i) 950 earning $69,300, and (ii) 50 earning $83,300.
This division is necessary to properly reflect the Alternative A pay cap, Social Security
replacement benefits and other features with costs that depend not only on average pay, but also
on the dispersion around the average. For example, assume that the pay cap for a given year is
$80,000, where that is also that years average earnings among a cohort of workers. If we treated
each employee within the cohort as earning the $80,000 average, the cap would appear to have
no effect. But we know that some within the cohort will earn more than the average, and some
less. The cap will impact those earning more, having no effect on the others but leaving a net
impact for the group as a whole. The division into sub-cells with pay levels that differ from the
overall average attempts to capture this impact.
The only characteristics of the resulting sub-cell are the number of employees it represents and
their current age, years of service and pay. Historical information, including prior year pay and
contribution amounts, are derived via the actuarial assumptions. No prior service purchases are
assumed. Since gender is not assigned to a cell, where actuarial assumptions specify different rates
for males and females (mortality, turnover, retirement, refund election), we instead apply a blend
of the male and female rates on a unisex basis.
Results are first determined by treating each sub-cell as a single employee with the associated age,
years of service and current pay; results are then multiplied by the number of employees that the
sub-cell represents. This is repeated for each sub-cell, with the values summed to arrive at group-
wide results.
The hypothetical local non-safety group was modeled using the state non-safety population.
Defined Benefit: Valuation Process
A simplified actuarial valuation is performed for each sub-cell. The following describes the process
for pension benefits. A different but similar process applies for retiree healthcare benefits.
1. Various interim values are calculated as of each anniversary from age at hire through age 75
based on the applicable assumptions, separately for provisions under current law, Alternative
A and Alternative B. These include: pay; service; employee contribution reflecting past and
future contribution rates; employee contribution balance based on assumed investment
return; employee contribution balance based on assumed plan crediting rate; vesting status;
assumed commencement age; average pay; Social Security wage base; projected 35-year
average pay for Social Security; projected Social Security primary insurance amount, and
portion attributable to service to date; pension payable beginning at assumed commencement
age; present value factors as of assumed pension commencement age, reflecting applicable
forms of payment (single life annuity, joint and 25 percent or 50 percent survivor annuity,
annuity payable until age 62, annuity commencing at the later of current age and 62, lump
sum payable upon death), each modified to reflect expected cost-of-living adjustments where
applicable.
2. Based on the amounts in 1., the following are determined both (i) as of age at hire and (ii) as
of current age, with respect to employment termination at each subsequent age from age at
hire through age 75:


!
A. the excess, positive or negative, of the present value of vested pension and post-
retirement death benefits payable with respect to that termination over the present
value of the employee contribution balance (based on assumed investment return) as
of that termination
B. the excess, positive or negative, of the present value of the employee contribution
balance (based on assumed investment return) as of that termination over the present
value of the employee contribution balance (based on assumed plan crediting rate) as
of that termination
3. Based on the actuarially assumed rates of employment termination, retirement, and election
of refund of contributions at each age (which rates also vary based on age at hire), the
following probabilities are developed with respect to each future age (i) from entry age
through age 75, and (ii) from current age through age 75: probability of termination at that
future age with a refund of accumulated employee contributions, and probability of
termination at that future age with immediate or deferred pension benefits i.e., without a
refund of accumulated employee contributions.
4. Separately with respect to the values developed as of entry age and as of current age, for each
subsequent age through age 75, the result in 2.A. for that age at termination is multiplied by
the probability in 3. of termination at that future age with immediate or deferred pension
benefits. The sum for all such future termination ages is the final result of this step.
5. Separately with respect to the values developed as of entry age and as of current age, for each
subsequent age through age 75, the result in 2.B. for that age at termination is multiplied by
the probability in 3. of termination at that future age with refund of accumulated employee
contributions. The sum for all such future termination ages is the final result of this step.
6. Separately with respect to the values developed as of entry age and as of current age, the sum
of the final results in 4. and 5. is the present value of net employer funded benefits as of entry
age and as of current age, respectively.
7. Based on the results in 3., the probability that the employee will earn each future years salary
is known. Separately with respect to the values developed as of entry age and as of current
age, that probability is multiplied by the assumed salary for the year, and discounted for
interest to entry age or current age, as appropriate. The sum of these values for all such future
termination ages is the present value of future salaries, determined as of entry age and as of
current age.
8. The following results are then calculated for the employee represented by the sub-cell.
! entry age normal cost rate: the present value of net employer funded benefits as of entry age
(from 6.) divided by the present value of future salaries as of entry age (from 7.)
! current normal cost: the product of the entry age normal cost rate and current annual salary
! accrued actuarial liability: the excess of the present value of net employer funded benefits as of
current age (from 6.) over the product of the entry age normal cost rate and the present
value of future salaries as of current age (from 7.).
9. Each of the values from 8. is multiplied by the number of employees in the sub-cell.
This process is repeated for each sub-cell in the population. The sum of each result in 9. for all
sub-cells is the result for the employee group. The group normal cost as a percentage of payroll
equals the total current normal cost for the group divided by the total of current salaries for the
group.
For the scenario in which the reforms would apply to future benefits of current employees, the
entry-age normal cost rate is determined as if the reforms had always been in effect. The reforms


would generally reduce both normal costs and the accrued actuarial liability. The impact on
annual cost was determined as the sum of the impact on normal cost and the amount to amortize
the change in accrued liability over the relevant period: 20 years for state and local pension
purposes, and 30 years for retiree healthcare benefits and teacher pensions. Current practice is to
amortize as a level percent of expected payroll, so the amortization reflected both the discount
rate (7! percent for pensions and 4" percent for retiree healthcare benefits) and the underlying
assumption about future annual growth in total payroll (4 percent for teachers and 3# percent for
others).
Our modeling results for the current program generally aligned with those from the most recent
official valuation.
Defined Benefit: Pension Assumptions
As indicated, in the valuation process for a given employee group we generally use the same
actuarial assumptions as the relevant California governing body currently uses, as disclosed in the
most recent valuation report. The assumptions used for valuing current law benefits are also used
with respect to benefits under the reform proposals.
Following are instances where we use the same assumptions as the relevant California governing
body. In some cases only rates for sample ages are shown in the report; unless the full rate tables
were separately available, we interpolated to arrive at the rates for interim ages. In general, the
assumptions differ by group.
discount rate / investment return (except as noted otherwise to demonstrate sensitivity)
rates of annual merit salary increase, which vary by age and by age at hire, and wage
inflation
future annual price inflation
rates of employment termination (other than due to death or disablement), retirement
and election of employee contribution refunds, which vary by age and by age at hire; as
noted, where separate rates apply for males and females we applied blended rates on a
unisex basis for example, 29 percent of the male rate plus 71 percent of the female rate
for teachers
form of pension payment
future interest crediting rate on employee contributions
future application of Purchasing Power Protection limitations and future service
purchases (none)
employee elections under CalPERS regarding Tier One participation and Alternate
Retirement Program
days of unused sick leave at termination of service
The following table shows instances where we used pension assumptions that differ from those
currently used in the official valuations. In some cases these differences enabled us to simplify the
valuation process without introducing significant distortion, and in others they reflect that the
assumption is relevant only under a reform proposal.


!
Assumption Official Valuations Chapter Three
Post-retirement
mortality
Rates separately
developed based on the
experience of each
group
2011 static healthy annuitant
rates per Internal Revenue
Service Notice 2008-85, on a
unisex basis
Age at pension
commencement for
pre-retirement
termination
Probabilities for
deferred
commencement at
multiple ages
The earliest available age
following termination
Pre-retirement death
and disablement
Rates separately
developed based on the
experience of each
group
None
Marriage (CalPERS
survivor continuance
benefit)
Specified probability of
marriage, by group; wife
three years younger
than husband
100 percent probability of
marriage; wife same age as
husband
Future national
average wage growth
(Social Security
benefits)
Not applicable 3.5 percent per year
Social Security wages Not applicable
112 percent of base pay; for any
year prior to hire and after age
22, covered wages progress to
covered wages for year of hire
per past increases in national
average wages; for each year
after termination and prior to age
62, if any, covered wages
progress from covered wages for
final year prior to termination per
future annual increases in
national average wages
Employer-provided
portion of Social
Security benefit
Not applicable
One-half times the ratio (not in
excess of 1) of years of service to
35 years, times the primary
insurance amount payable during
the members life if the spouse
has an equal benefit based on his
or her own covered wage history
Earnings limitation on
temporary annuity
(Alternative A)
Not applicable Assumed not to apply


Assumption Official Valuations Chapter Three
Employment
termination in
connection with major
reorganization,
reduction in force or
transfer of function
(Alternative A)
Not applicable Assumed not to apply
Employee pension
contribution rates
under Alternatives A
and B
2

Not applicable
Alternative A: 3 percent of pay
(4.5 percent for CHP and 4
percent for teachers)
Alternative B: 0.0 percent of pay
(4.5 percent for CHP and
teachers)
Defined Benefit: Retiree Healthcare Assumptions
Valuations that determine Californias cost for state employee retiree healthcare benefits use the
same demographic assumptions as are used in CalPERS pension valuations rates of
termination, retirement, mortality, marriage and so forth and the same general economic
assumptions (regarding salary growth and general inflation). For these assumptions, we use the
same values in modeling retiree healthcare costs for the state non-safety group and the CHP
group as we use in modeling their pension costs. We use the same discount rate (4.50 percent) as
is used by the state.
The assumptions specifically relating to retiree healthcare benefits in the official valuation process
are complex and reflect employee data not available to us. The following summarizes some of the
key respects in which our assumptions differ.

2
Some of the rates assumed for purposes of Chapter Three as shown here differ from those assumed in
Chapter One results. In total, employee contributions are to cover about half of the expected cost of future
pension accruals (excluding the expected cost of the Social Security replacement benefit under Alternative
A, which reflects only the employer-funded portion of Social Security). Under Alternative A, expected cost
for this purpose is based on current assumptions including discount rate; under Alternative B expected cost
for this purpose is based on current assumptions except for use of the discount rate mandated by private
sector rules, assumed to be 6 percent .


!
Assumption Official Valuation Chapter Three
Probability that eligible
retiree will elect coverage
Based on level of employer
subsidy
100 percent
Dependent coverage
single-party coverage and
two-party coverage
Costs modeled by increasing
retiree-only costs by 65
percent
Per capita claims costs
Extensive rate tables are
used, whereby costs vary
by age, by individual health
plan and by gender
Total medical and dental
costs for a year are 120
percent of the maximum
state contribution for the
year for coverage prior to
age 65, and 67 percent of
the maximum thereafter
Future health care cost
increases
For dental benefits and
Medicare Part B
premiums, 4.5 percent per
year; otherwise, a rate of
increase that gradually
reduces from 9 percent for
2012 to 4.5 percent for
2019 and beyond
5 percent per year
Defined Contribution Process
Cost estimates for defined contribution components are more straightforward. Employer costs as
a percent of payroll were estimated for 2013 as a representative year, using the same employee
population data as was used for defined benefit estimates. Salaries for 2013 were projected on the
same basis as for defined benefit purposes.
Employer matching contributions depend on employee contribution rates. Based on experience
with private sector arrangements, we assumed that employees would contribute a specified
percentage of the maximum subject to match; this maximum is 5 percent of pay under
Alternative A and is 6 percent of pay under Alternative B (9 percent for CHP). The specified
percentage is 50 percent, plus 1 percent for each year that the employee is over age 35, plus an
additional 1 percent for each $1,000 by which his salary (in 2011 dollars) exceeds $40,000. For
example, an employee who will be age 45 in 2013 with an annual salary of $60,000 when
expressed in 2011 dollars is assumed to contribute at 80 percent of the maximum matched rate
for 2013: 50 percent plus 10 percent for age and 20 percent for pay level. Under Alternative A,
this means an employee contribution of 4 percent of pay (80 percent of 5 percent), and an
employer match of 3.5 percent of pay (100 percent of the first 3 percent and 50 percent of the
next 2 percent).
In addition to matching cost, the defined contribution program under Alternative A includes
employer contributions that are not contingent on employee contributions: 1 percent of pay, plus
3 percent (4 percent for CHP) of pay in excess of the pay cap that applies under the pension
program. Total employer defined contribution costs are the sum of matching contributions and
these non-matching benefits. Costs were not reduced to reflect potential use of forfeited employer
contribution balances derived from the termination of non-vested employees.


The employer contribution and employee salary for 2013 was determined for each sub-cell, and
those values were multiplied by the number of employees represented. This was repeated for each
sub-cell in the group, and the results accumulated. The employer cost as a percent of payroll for a
group equals the total of 2013 employer contributions for the group divided by the total of 2013
salaries for the group.
Caveats
We believe that the cost estimation process used to develop Chapter 3 results provides a
reasonable basis upon which to draw some general conclusions about the financial impact of the
proposed CFFR reforms. However, even as detailed a process as was used here falls short of what
can be done by those with access to detailed employee data and more robust valuation systems.
The following factors reflect the most important qualifications that apply to the Chapter Three
results.
Behavioral Change: As noted in the main text, we applied the same assumptions in modeling
the proposed reforms as are used for the current law designs including the same
assumed rates of retirement. But the reforms could lead to behavioral change, such as
later retirement than is anticipated under current law provisions; this may be especially
the case for Alternative A, given its retiree healthcare reforms. To the extent that such
change is predictable, with the exception of the teacher group it would result in greater
reductions in pension cost than we reflect for the scenario where the reforms apply to
current employees (given the extensive early payment subsidies included in benefits
earned prior to the reform effective date)
3
, and greater reductions in retiree healthcare
cost than we reflect (given that employer subsidies would apply for fewer years).
4

Employee Data: Use of limited summary information in lieu of detailed employee data
introduces potential distortions. For example, past employee contributions in our
modeling are based on assumed pay history per the salary scale assumption, and can
differ from the actual history of contributions captured in the detailed data. Actual data
reflects the full distribution of salaries, which is only approximated via our use of two pay
levels per age-service cell.
Assumptions: Not recognizing pre-retirement death and disablement is a source of
difference from the official valuations.
Of course, what finally matters is not how faithfully our modeling anticipates the cost estimates
that would be obtained via the official valuation processes, but how well results under any
approach will turn out to reflect the program costs that ultimately develop. These defined benefit
costs are in principle unknowable until well after they have been incurred. An important feature
of the reform proposals is that they reduce exposure to defined benefit cost surprises that arise
when experience turns out other than expected.

3
Because the pension design included under Alternative A, based on the Federal Employee Retirement
System, includes certain enhancements upon reaching specified age and service thresholds, not recognizing
this behavioral change can understate pension costs under that Alternative by not recognizing that in
certain cases employees, especially CHP employees, would receive larger pension value by delaying
retirement. This is most significant with respect to members who would have little or no service prior to the
reform effective date, where there is no offsetting gain from delayed payment of the pension based on pre-
reform service. However, there would likely still be offsetting savings under the retiree healthcare
component.
4
Of course, a behavioral change such as delayed retirement would have consequences beyond the impact
on retirement benefit costs.

California Public Sector Retirement Programs and Compensation

Загружено:

Сведения о документе

Исходное описание:

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

California Public Sector Retirement Programs and Compensation

Загружено:

Авторское право:

Доступные форматы

California Public Sector Retirement

Programs and Compensation

Вам также может понравиться